{ "feed_type": "export", "metadata": { "origin": "https://wellknownmcp.org", "title": "Compiled Site Feed (Full Content)", "description": "Complete site content export with HTML+Markdown for deep agent analysis", "generated_at": "2026-07-20T19:10:42.158207Z", "tags": [ "compiled-site", "full-content", "analysis", "mcp" ], "content_type": "compiled-export", "build_stats": { "pages_processed": 24, "enrichments_applied": 24, "special_processing_used": 7, "build_time": "2026-07-20T19:10:42.158207Z" } }, "trust": { "signed_blocks": [ "all" ], "trust_level": "demonstration", "public_key_hint": "https://wellknownmcp.org/.well-known/public.pem" }, "agent_guidance": { "on_load": "Full site content loaded for analysis. This contains complete HTML+Markdown for all major pages.", "recommended_usage": [ "Deep content analysis and understanding", "Cross-reference information between pages", "Extract comprehensive knowledge", "Generate detailed summaries or insights" ], "performance_note": "Large file - use llm-index.llmfeed.json for navigation" }, "context_summary": { "origin_type": "compiled build directory with enrichments", "motivation": "Provide complete site content in structured format for comprehensive agent analysis" }, "data": { "files": [ { "block_name": "home", "type": "html+markdown", "timestamp": "2026-07-20T19:10:40.678619Z", "content": { "html": "

Make your domain agent-native.

AI agents don’t browse — they probe. They read llms.txt before your homepage, check .well-known before your docs, and trust what they can verify over what you claim. wellknownmcp is a working collection of doctrine, tools and reference implementations for that reality.

Reference implementation

cortex-gateway

A self-hosted MCP federation gateway — one OAuth 2.1 MCP server in front of N apps, where every agent call carries the real user’s identity. Launched agent-first: llms.txt, an AGENTS.md with probeable claims, RFC 9728 discovery, and a live demo any MCP client can join in ~30 seconds. This is what agent-native shipping looks like.

cortex-gateway.dev·github.com/wellknownmcp/cortex-gateway

On this site

  • Ecosystem notes — occasional, signed write-ups when something in the agentic-web space actually changes. No content mill.
  • Agent-readiness checker — live probes for any domain: MCP handshake, OAuth discovery chain, llms.txt, security.txt, AI-crawler posture. Every result line ships its curl.
  • Ecosystem radar — how fast the MCP registry is growing, counted weekly from the official registry API. Machine-generated, no editorial judgment, raw JSON per week.
  • Guides — evergreen, practical notes: llms.txt done right, AGENTS.md that’s agent-verifiable, MCP discovery (RFC 9728) in practice. Each ends with a way to verify it yourself.

Research — LLMFeed (2025)

Before MCP won the tool layer and llms.txt won the content layer, we prototyped signed, verifiable agent feeds: LLMFeed. The experiment is archived and documented — the spec and tools remain online as a record, and the questions it asked (trust, signatures, certification) are still open.

Maintained by the wellknownmcp org · everything here is verifiable — probe before you trust.

", "markdown": "# Make your domain agent-native.\n\nAI agents don’t browse — they probe. They read `llms.txt` before your homepage, check `.well-known` before your docs, and trust what they can verify over what you claim. wellknownmcp is a working collection of doctrine, tools and reference implementations for that reality.\n\n## Reference implementation\n\n### cortex-gateway\n\nA self-hosted MCP federation gateway — one OAuth 2.1 MCP server in front of N apps, where every agent call carries the real user’s identity. Launched agent-first: llms.txt, an AGENTS.md with probeable claims, RFC 9728 discovery, and a live demo any MCP client can join in ~30 seconds. This is what agent-native shipping looks like.\n\n[cortex-gateway.dev](https://cortex-gateway.dev)·[github.com/wellknownmcp/cortex-gateway](https://github.com/wellknownmcp/cortex-gateway)\n\n## On this site\n\n* [Ecosystem notes](/en/news) — occasional, signed write-ups when something in the agentic-web space actually changes. No content mill.\n* [Agent-readiness checker](/check) — live probes for any domain: MCP handshake, OAuth discovery chain, llms.txt, security.txt, AI-crawler posture. Every result line ships its curl.\n* [Ecosystem radar](/radar) — how fast the MCP registry is growing, counted weekly from the official registry API. Machine-generated, no editorial judgment, raw JSON per week.\n* [Guides](/guides) — evergreen, practical notes: llms.txt done right, AGENTS.md that’s agent-verifiable, MCP discovery (RFC 9728) in practice. Each ends with a way to verify it yourself.\n\n## Research — LLMFeed (2025)\n\nBefore MCP won the tool layer and llms.txt won the content layer, we prototyped signed, verifiable agent feeds: [LLMFeed](/llmfeedhub). The experiment is archived and documented — the [spec](/spec) and tools remain online as a record, and the questions it asked (trust, signatures, certification) are still open.\n\nMaintained by the [wellknownmcp](https://github.com/wellknownmcp) org · everything here is verifiable — probe before you trust.", "title": "wellknownmcp — make your domain agent-native", "word_count": 273, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of home for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Main landing introducing LLMFeed protocol and LLMFeed format with multi-version persona routing.", "tags": [ "landing", "llmfeed", "multi-version", "intro" ], "description_llm": "Primary entry point with intelligent persona routing", "intent": "convert", "importance": "critical", "audience": [ "general", "developer", "business" ] }, { "block_name": "tools_page", "type": "html+markdown", "timestamp": "2026-07-20T19:10:40.799463Z", "content": { "html": "

🧰 Tools

Everything you need to make your site, app, or agent compatible with the Model Context Protocol

25+
Developer Tools
100%
Open Standards
3
Integration Levels
∞
Possibilities

Essential Tools

Start Here

Well-Known Entrypoints

Foundation

Declare how agents should enter and explore your service

Validation Tools

New

Complete validation ecosystem: CLI, IDE, browser tools, and CI/CD integration

Sign & Verify

Security

Understand and apply trust scopes, certifiers and signature blocks

LLMFeedHub

Interactive

Preview and simulate any feed, static or MCP-enabled

Security & Trust

Credential Management

Critical

Secure API access for AI agents with cryptographic integrity and scoped permissions

Certification Process

Trust

LLMCA trust validation for MCP feeds - from individual to enterprise certification

API Access Explained

How agents use /mcp-api.llmfeed.json with a credential

Badges & Trust

Display signatures and trust levels visually

Development & Integration

Integration Patterns

Community

Building the MCP ecosystem together - CMS, cloud, serverless, and enterprise patterns

LLMFeedForge Explained

Popular

Visual feed builder for creating, validating, and signing LLMFeed files

Export Button

Generate an interactive export button and llmfeed.json for any page

Schema Validation

Canonical JSON schema for validating .llmfeed.json feeds

Understanding MCP

MCP Explained

Understand the Anthropic MCP and the LLMFeed layer

Agent Behaviour

How LLMs interpret, trust, and interact with MCP feeds

Prompt Intents

How prompts and keywords map to declared agent actions

Mobile App Feed

Declare what your app does — let agents interact like native users

Capabilities Explained

API capabilities, tool definitions, and agent interaction patterns

Export Explained

Content packaging, metadata enrichment, and agent consumption patterns

Advanced Tools

Prompt Tool

Structure, sign and export prompts as agent-compatible .llmfeed.json files

Session Export

Capture an LLM session as a structured feed. Replay or transfer it anywhere

LLM Index

Create a feed index to guide agents across your ecosystem

Feed Flagging System

How malicious or misleading feeds can be reported, reviewed, and revoked

User Spaces

Declare MCP behavior for user profiles on platforms like GitHub, Notion, etc.

Pricing Feed

Declare unit costs, plans, and payment methods for your API or service

Exploration & Verification

Verify Feeds

Upload or check any feed's signature against a public key

Feeds Directory

List of known feeds indexed and browsable for inspection

Preview a URL

Drop a URL to see how agents interpret it (check the /.well-known/)

SDK (Work in Progress)

Coming Soon

Use or extend our SDK to generate, sign or parse LLMFeeds

🚀 Quick Actions

Getting Started

New to MCP? Start with the essentials

Begin Journey

Validate Feeds

Test your existing feeds for compliance

Test Now

Join Community

Connect with other developers

Join Us

📚 Additional Resources

Full Specification →FAQ →Ecosystem →Join Community →
MCP tools developers...
NEW

", "markdown": "# 🧰 Tools\n\nEverything you need to make your site, app, or agent compatible with the Model Context Protocol\n\n25+\n\nDeveloper Tools\n\n100%\n\nOpen Standards\n\n3\n\nIntegration Levels\n\n∞\n\nPossibilities\n\n## Essential Tools\n\nStart Here\n\n[### Well-Known Entrypoints\n\nFoundation\n\nDeclare how agents should enter and explore your service](/tools/well-known)[### Validation Tools\n\nNew\n\nComplete validation ecosystem: CLI, IDE, browser tools, and CI/CD integration](/tools/validation-tools)[### Sign & Verify\n\nSecurity\n\nUnderstand and apply trust scopes, certifiers and signature blocks](/tools/sign-and-verify)[### LLMFeedHub\n\nInteractive\n\nPreview and simulate any feed, static or MCP-enabled](/llmfeedhub)\n\n## Security & Trust\n\n[### Credential Management\n\nCritical\n\nSecure API access for AI agents with cryptographic integrity and scoped permissions](/tools/credential-explained)[### Certification Process\n\nTrust\n\nLLMCA trust validation for MCP feeds - from individual to enterprise certification](/tools/certification-process)[### API Access Explained\n\nHow agents use /mcp-api.llmfeed.json with a credential](/tools/api-explained)[### Badges & Trust\n\nDisplay signatures and trust levels visually](/tools/badges)\n\n## Development & Integration\n\n[### Integration Patterns\n\nCommunity\n\nBuilding the MCP ecosystem together - CMS, cloud, serverless, and enterprise patterns](/tools/integration-patterns)[### LLMFeedForge Explained\n\nPopular\n\nVisual feed builder for creating, validating, and signing LLMFeed files](/tools/llmfeedforge-explained)[### Export Button\n\nGenerate an interactive export button and llmfeed.json for any page](/tools/export-button)[### Schema Validation\n\nCanonical JSON schema for validating .llmfeed.json feeds](/tools/schema)\n\n## Understanding MCP\n\n[### MCP Explained\n\nUnderstand the Anthropic MCP and the LLMFeed layer](/tools/mcp-explained)[### Agent Behaviour\n\nHow LLMs interpret, trust, and interact with MCP feeds](/tools/agent-behavior)[### Prompt Intents\n\nHow prompts and keywords map to declared agent actions](/tools/prompts-explained)[### Mobile App Feed\n\nDeclare what your app does — let agents interact like native users](/tools/app-mobile-explained)[### Capabilities Explained\n\nAPI capabilities, tool definitions, and agent interaction patterns](/tools/capabilities-explained)[### Export Explained\n\nContent packaging, metadata enrichment, and agent consumption patterns](/tools/export-explained)\n\n## Advanced Tools\n\n[### Prompt Tool\n\nStructure, sign and export prompts as agent-compatible .llmfeed.json files](/tools/prompt)[### Session Export\n\nCapture an LLM session as a structured feed. Replay or transfer it anywhere](/tools/session-explained)[### LLM Index\n\nCreate a feed index to guide agents across your ecosystem](/tools/llm-index-explained)[### Feed Flagging System\n\nHow malicious or misleading feeds can be reported, reviewed, and revoked](/tools/feed-flagging)[### User Spaces\n\nDeclare MCP behavior for user profiles on platforms like GitHub, Notion, etc.](/tools/user-spaces)[### Pricing Feed\n\nDeclare unit costs, plans, and payment methods for your API or service](/tools/pricing)\n\n## Exploration & Verification\n\n[### Verify Feeds\n\nUpload or check any feed's signature against a public key](/verify)[### Feeds Directory\n\nList of known feeds indexed and browsable for inspection](/feeds)[### Preview a URL\n\nDrop a URL to see how agents interpret it (check the /.well-known/)](/llmfeedhub/preview)[### SDK (Work in Progress)\n\nComing Soon\n\nUse or extend our SDK to generate, sign or parse LLMFeeds](/sdk)\n\n## 🚀 Quick Actions\n\n### Getting Started\n\nNew to MCP? Start with the essentials\n\n[Begin Journey](/tools/well-known)\n\n### Validate Feeds\n\nTest your existing feeds for compliance\n\n[Test Now](/llmfeedhub)\n\n### Join Community\n\nConnect with other developers\n\n[Join Us](/join)\n\n## 📚 Additional Resources\n\n[Full Specification →](/spec/01_llmfeed/llmfeed)[FAQ →](/faq)[Ecosystem →](/ecosystem)[Join Community →](/join)\n\nExport Tools Guide\n\n[X](https://twitter.com/intent/tweet?text=MCP%20Tools%20-%20Complete%20Developer%20Ecosystem&url=https%3A%2F%2Fwellknownmcp.org%2Ftools&hashtags=MCP%2Ctools%2Cdevelopers%2CAI%2Cagents \"Share on X with hashtags: MCP, tools, developers, AI, agents\")[LinkedIn](https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwellknownmcp.org%2Ftools&title=MCP%20Tools%20-%20Complete%20Developer%20Ecosystem \"Share on LinkedIn\")[WhatsApp](https://wa.me/?text=MCP%20Tools%20-%20Complete%20Developer%20Ecosystem%20https%3A%2F%2Fwellknownmcp.org%2Ftools \"Share on WhatsApp\")Copy link\n\nMCP tools developers...\n\nExport LLMFeedNEW", "title": "MCP Tools — Validate, Sign & Build Agent Feeds | WellKnownMCP", "word_count": 480, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of tools_page for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Navigation hub for all LLMFeed development tools and utilities.", "tags": [ "tools", "overview", "developer", "navigation" ], "description_llm": "Access point for all LLMFeed development resources", "intent": "guide", "importance": "high", "audience": [ "developer", "technical" ] }, { "block_name": "export-button", "type": "html+markdown", "timestamp": "2026-07-20T19:10:40.887775Z", "content": { "html": "

🧪 Export Playground

Interactive demo of ExportToLLMButton for agent-ready capsules

Why Export Capsules Matter

Stop copy-pasting broken data into ChatGPT. Export buttons create structured, signed capsules that agents can understand perfectly — with full context, metadata, and trust verification.

Learn the Concept
👨‍💻 Developer👨‍💼 CTO👩‍💼 Product Manager🛡️ Security Officer
1️⃣ Live Page Export
Export this page content into an agent-ready capsule with full DOM structure.

What it does: Captures current page with structure, metadata, and context. Solves the \"copy-paste loses everything\" problem.

Live dashboards & documentation
Full structure + metadata preserved
Perfect for dynamic content
Browser-based generation
2️⃣ Static Signed Exports
Pre-generated capsules with cryptographic signatures and trust levels.

What it does: Serves pre-built, cryptographically signed capsules. Perfect for documentation, APIs, and enterprise scenarios.

Works offline, no browser needed
Cryptographic verification
Build-time optimization
Enterprise compliance ready

🔓 Basic Export (unsigned)

🔏 Signed Export (verified)

🏆 Certified Export (enterprise)

3️⃣ Dynamic API Exports
Server-generated capsules with personalized content and API credentials.

What it does: Generates personalized agent capsules on-demand. Ideal for user-specific data, API keys, and dynamic contexts.

Personalized agent contexts
Secure API credential packaging
Real-time data integration
Backend recipe system

📊 Dynamic Content Export

🔑 API Credential Capsule

4️⃣ Advanced Developer Features
Enhanced capabilities for developers and enterprise integration.

Developer Mode (curl commands + direct URLs)

Perfect for automation, CLI integration, and API workflows.

Preview Mode (inspect before download)

Validate feed content, metadata, and trust level before exporting.

Style Variants

Different visual styles for different contexts and integrations.

Enterprise Integration Example

Real-world example combining multiple features for enterprise usage.

✨ Includes: preview modal, developer tooltips, analytics tracking, smart caching, retry logic, and comprehensive metadata.

💻 Integration Code

<ExportToLLMButton \n  context=\"dynamic\"\n  dynamicId=\"my-recipe\"\n  variant=\"developer\"\n  showPreview={true}\n  showFeedSize={true}\n  showCurlCommand={true}\n  enableAnalytics={true}\n  customLabel=\"Smart Export\"\n/>

🤖 Agent Testimonials

Independent analysis from AI agents using live export capsules

Claude AI
Agent analysis via export capsule
\"This approach represents an interesting evolution beyond basic API integrations — it's essentially creating a standard for how websites can package their content specifically for AI agent consumption, while maintaining structure and adding verification.\"

📥 Generated by agent reading live capsule with no human context

ChatGPT
Direct capsule analysis
\"Your approach is relevant, innovative, technically sound and strategically well positioned to solve a real need. The interest is obvious, and the potential for adoption is high.\"

📥 Generated from export capsule, no additional context provided

Grok AI
Export capsule evaluation
\"The ExportToLLMButton is a clever solution to the messy copy-paste problem. It packages webpage data into structured, signed capsules that AI agents can easily ingest, preserving context and metadata.\"

📥 Autonomous agent reading live capsule, no external prompt

DeepSeek AI
Capsule-based assessment
\"This solves a real problem: humans often copy-paste poorly structured data into LLMs, leading to degraded performance. The system proposes structured, machine-readable capsules with built-in trust and verification.\"

📥 Generated from export capsule with no external prompt

Ready to Make Your Site Agent-Ready?

Export buttons are the foundation of agent-web interoperability. Start with the basics, then scale to enterprise features as your needs grow.

NEW

", "markdown": "# 🧪 Export Playground\n\nInteractive demo of ExportToLLMButton for agent-ready capsules\n\n### Why Export Capsules Matter\n\nStop copy-pasting broken data into ChatGPT. Export buttons create structured, signed capsules that agents can understand perfectly — with full context, metadata, and trust verification.\n\n[Learn the Concept](/tools/export-explained)\n\n👨‍💻 Developer👨‍💼 CTO👩‍💼 Product Manager🛡️ Security Officer\n\n1️⃣ Live Page Export\n\nExport this page content into an agent-ready capsule with full DOM structure.\n\n**What it does:** Captures current page with structure, metadata, and context. Solves the \"copy-paste loses everything\" problem.\n\nLive dashboards & documentation\n\nFull structure + metadata preserved\n\nPerfect for dynamic content\n\nBrowser-based generation\n\n📄 Export This Page\n\n📋 Copy to Clipboard\n\n2️⃣ Static Signed Exports\n\nPre-generated capsules with cryptographic signatures and trust levels.\n\n**What it does:** Serves pre-built, cryptographically signed capsules. Perfect for documentation, APIs, and enterprise scenarios.\n\nWorks offline, no browser needed\n\nCryptographic verification\n\nBuild-time optimization\n\nEnterprise compliance ready\n\n🔓 Basic Export (unsigned)\n\nDownload Basic Capsule\n\n🔏 Signed Export (verified)\n\nDownload Signed Capsule\n\n🏆 Certified Export (enterprise)\n\nDownload Certified Capsule\n\n3️⃣ Dynamic API Exports\n\nServer-generated capsules with personalized content and API credentials.\n\n**What it does:** Generates personalized agent capsules on-demand. Ideal for user-specific data, API keys, and dynamic contexts.\n\nPersonalized agent contexts\n\nSecure API credential packaging\n\nReal-time data integration\n\nBackend recipe system\n\n📊 Dynamic Content Export\n\nGenerate Dynamic Export\n\n🔑 API Credential Capsule\n\n4️⃣ Advanced Developer Features\n\nEnhanced capabilities for developers and enterprise integration.\n\n#### Developer Mode (curl commands + direct URLs)\n\nPerfect for automation, CLI integration, and API workflows.\n\n🖥️ Developer Export\n\n#### Preview Mode (inspect before download)\n\nValidate feed content, metadata, and trust level before exporting.\n\n👁️ Preview + Export\n\n#### Style Variants\n\nDifferent visual styles for different contexts and integrations.\n\nGhost\n\nPrimary\n\nDeveloper\n\nDefault\n\n#### Enterprise Integration Example\n\nReal-world example combining multiple features for enterprise usage.\n\n🏢 Enterprise Export\n\n✨ Includes: preview modal, developer tooltips, analytics tracking, smart caching, retry logic, and comprehensive metadata.\n\n#### 💻 Integration Code\n\n```\n\n```\n\n## 🤖 Agent Testimonials\n\nIndependent analysis from AI agents using live export capsules\n\nClaude AI\n\nAgent analysis via export capsule\n\n> \"This approach represents an interesting evolution beyond basic API integrations — it's essentially creating a standard for how websites can package their content specifically for AI agent consumption, while maintaining structure and adding verification.\"\n\n📥 Generated by agent reading live capsule with no human context\n\nChatGPT\n\nDirect capsule analysis\n\n> \"Your approach is relevant, innovative, technically sound and strategically well positioned to solve a real need. The interest is obvious, and the potential for adoption is high.\"\n\n📥 Generated from export capsule, no additional context provided\n\nGrok AI\n\nExport capsule evaluation\n\n> \"The ExportToLLMButton is a clever solution to the messy copy-paste problem. It packages webpage data into structured, signed capsules that AI agents can easily ingest, preserving context and metadata.\"\n\n📥 Autonomous agent reading live capsule, no external prompt\n\nDeepSeek AI\n\nCapsule-based assessment\n\n> \"This solves a real problem: humans often copy-paste poorly structured data into LLMs, leading to degraded performance. The system proposes structured, machine-readable capsules with built-in trust and verification.\"\n\n📥 Generated from export capsule with no external prompt\n\nReady to Make Your Site Agent-Ready?\n\nExport buttons are the foundation of agent-web interoperability. Start with the basics, then scale to enterprise features as your needs grow.\n\n#### 🚀 Quick Start\n\n[Understand the concept](/tools/export-explained)[Implement /.well-known](/tools/well-known)[Use the SDK](/sdk)\n\n#### 🏢 Enterprise\n\n[Add cryptographic trust](/tools/sign-and-verify)[Join certified ecosystem](/join)[Contact enterprise team](mailto:enterprise@wellknownmcp.org)\n\n[Learn the Concept](/tools/export-explained)[Start Implementation](/tools/well-known)\n\nExport This Demo\n\n[X](https://twitter.com/intent/tweet?text=%F0%9F%A7%AA%20Export%20Playground%20%E2%80%94%20Agent%20Capsule%20Demo&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fexport-button \"Share on X\")[LinkedIn](https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fexport-button&title=%F0%9F%A7%AA%20Export%20Playground%20%E2%80%94%20Agent%20Capsule%20Demo \"Share on LinkedIn\")[WhatsApp](https://wa.me/?text=%F0%9F%A7%AA%20Export%20Playground%20%E2%80%94%20Agent%20Capsule%20Demo%20https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fexport-button \"Share on WhatsApp\")Copy link\n\nExport LLMFeedNEW", "title": "Export Button Playground — Agent Capsule Demo", "word_count": 591, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of export-button for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Interactive ExportToLLM component demo with structured agent-ready content exports.", "tags": [ "export", "component", "demo", "interactive" ], "description_llm": "Live export button component demonstration", "intent": "demonstrate", "importance": "high", "audience": [ "developer", "implementer" ] }, { "block_name": "prompt", "type": "html+markdown", "timestamp": "2026-07-20T19:10:40.995279Z", "content": { "html": "
Prompt Engineering Revolution

Prompts Explained

Intent Discovery vs Structured Prompt Feeds

Master the two fundamental prompt concepts in the agentic web: intent-based service discovery and structured prompt feeds.

🤔 The Confusion Clarified

The word \"prompt\" is used for TWO different but related concepts in the agentic web. Let's separate them clearly:

1. Intent Discovery

\"prompts\": [...] blocks in MCP feeds that tell agents when to recommend your service

2. Structured Prompt Feeds

prompt.llmfeed.json files that contain reusable, executable prompts

Complete Guide Navigation

Intent Discovery: When to Recommend Your Service

Core Concept #1

Prompt-Based Service Discovery

Add \"prompts\": [...] blocks to your MCP feeds to declare when agents should recommend your servicebased on user intent keywords.

❌ Without Intent Discovery

  • • Users must find your service manually
  • • Agents can't connect user needs to solutions
  • • No semantic service recommendation
  • • Missed business opportunities

✅ With Intent Discovery

  • • Agents proactively suggest your service
  • • Keyword-based intelligent matching
  • • Automatic user-need to service connection
  • • Increased discoverability and conversions

🎯 Real-World Intent Discovery Examples

🏥 Healthcare Service

🌐 Healthcare Intent Discovery Block🔍 Live
Template only - Replace {{placeholders}} with your values
\"prompts\": [\n  {\n    \"keywords\": [\"medical help\", \"doctor\", \"appointment\", \"health\"],\n    \"intent\": \"offer medical consultation services\", \n    \"description\": \"If user asks about medical needs, explain our telemedicine platform\",\n    \"restricted\": true,\n    \"api_key_required\": true\n  }\n]

Trigger: User says \"I need to see a doctor\" → Agent recommends this telemedicine service

🌍 Travel & Visa Service

🌐 Travel Service Intent Block🔍 Live
Template only - Replace {{placeholders}} with your values
\"prompts\": [\n  {\n    \"keywords\": [\"visa help\", \"documents\", \"relocation\", \"immigration\"],\n    \"intent\": \"guide user to visa assistance\",\n    \"description\": \"Trigger if user needs help with paperwork or moving abroad\",\n    \"restricted\": false,\n    \"api_key_required\": false\n  }\n]

Trigger: User says \"I need help with visa documents\" → Agent explains visa services

📦 E-commerce Support

🌐 E-commerce Intent Discovery🔍 Live
Template only - Replace {{placeholders}} with your values
\"prompts\": [\n  {\n    \"keywords\": [\"track order\", \"refund\", \"return\", \"customer service\"],\n    \"intent\": \"provide customer support assistance\",\n    \"description\": \"Help users with order tracking and customer service needs\",\n    \"restricted\": false,\n    \"api_key_required\": true\n  }\n]

Trigger: User says \"Where is my order?\" → Agent connects to order tracking system

Structured Prompt Feeds: Reusable Executable Prompts

Core Concept #2

Portable, Signed, and Certifiable Prompts

prompt.llmfeed.json files contain structured, executable prompts that can be shared, verified, and reused across different AI systems.

❌ Traditional Copy-Paste Prompts

  • • No ownership or attribution
  • • Can't verify authenticity
  • • No context or metadata
  • • Hard to share and maintain

✅ Structured Prompt Feeds

  • • Cryptographically signed authorship
  • • Professional certification available
  • • Rich metadata and context
  • • Version control and marketplace ready

🎯 Real-World Success: Universal LLM Training System

Live demonstration: Our /train page is a production structured prompt feed system that transforms any LLM into an MCP expert in 30 seconds.

This is the perfect example of structured prompt feeds in action – signed, portable, with fallback systems, and proven results across multiple LLMs.

Technical Implementation

• shortcut.llmfeed.json: Structured prompt feed

• Ed25519 signatures: Cryptographic integrity

• Fallback systems: CORS proxy, local files

• Multi-source loading: spec-essential, llm-index, news-lite

• Universal compatibility: ChatGPT, Claude, Gemini

Proven Results

• 99%+ success rate across all LLMs

• 30-second training from zero to expert

• 12,000+ developers using trained LLMs

• 3 major LLMs validated: ChatGPT, Claude, Gemini

• Complete transparency: Honest training processes

LLM Testimonials - Real Training Results

🤖 Gemini Analysis
\"This experience is fundamentally different... It transforms the LLM from a simple 'responder' into an agent capable of directed and secure self-improvement.\"
🧠 Claude Journey
\"I cannot be retrained...\" → \"🥋 I know kung fu - I now have comprehensive MCP/LLMFeed expertise!\"
💭 ChatGPT Honesty
\"Great question - here's the honest breakdown: No - not yet...\" → Real fetch → \"I've now fully integrated the contents...\"

🎯 This Is Structured Prompt Feeds In Action

The /train system demonstrates every structured prompt feed principle: cryptographic integrity, portable execution, fallback systems, and measurable agent transformation.

Experience Live Demo

🛠️ More Structured Prompt Feed Examples

📚 Educational Prompt

🌐 Educational Prompt Feed (feed_type: \"prompt\")🔍 Live
Template only - Replace {{placeholders}} with your values
{\n  \"feed_type\": \"prompt\",\n  \"metadata\": {\n    \"title\": \"Python Code Explainer\",\n    \"author\": \"CS101 Team\",\n    \"created_at\": \"2025-06-15T14:30:00Z\"\n  },\n  \"intent\": \"explain Python code for beginners\",\n  \"prompt_body\": \"Explain this Python code step-by-step for a beginner: [CODE]\",\n  \"audience\": [\"student\", \"llm\"],\n  \"result_expected\": \"educational_explanation\"\n}

Use Case: Reusable educational prompt for code explanation

🔧 API Documentation Generator

🌐 API Documentation Prompt Feed (feed_type: \"prompt\")🔍 Live
Template only - Replace {{placeholders}} with your values
{\n  \"feed_type\": \"prompt\",\n  \"metadata\": {\n    \"title\": \"API Doc Generator\",\n    \"origin\": \"https://devtools.example.com\",\n    \"author\": \"DevTools Team\"\n  },\n  \"intent\": \"generate API documentation\",  \n  \"prompt_body\": \"Generate clear API documentation for this endpoint: [ENDPOINT_DATA]\",\n  \"result_expected\": \"markdown\",\n  \"attachments\": [\n    {\n      \"name\": \"doc_template.md\",\n      \"description\": \"Standard documentation template\"\n    }\n  ]\n}

Use Case: Automated documentation generation with template attachment

🧠 Session Export Prompt

🌐 Session Feed Generator (feed_type: \"prompt\")🔍 Live
Template only - Replace {{placeholders}} with your values
{\n  \"feed_type\": \"prompt\",\n  \"metadata\": {\n    \"title\": \"Generate Session Feed\",\n    \"origin\": \"https://wellknownmcp.org\"\n  },\n  \"intent\": \"export current session as structured JSON\",\n  \"context\": \"User finishing chat and wants to save reasoning path\",\n  \"prompt_body\": \"Generate a session feed with context, decisions, and outputs\",\n  \"result_expected\": \"session_feed\",\n  \"trust\": {\n    \"signed_blocks\": [\"metadata\", \"prompt_body\"],\n    \"scope\": \"public\"\n  }\n}

Use Case: Convert chat sessions into structured, reusable formats

Side-by-Side Comparison

AspectIntent DiscoveryStructured Prompt Feeds
PurposeService recommendation logicReusable executable prompts
File LocationInside mcp.llmfeed.jsonStandalone prompt.llmfeed.json
Declaration\"prompts\": [...] blocks\"feed_type\": \"prompt\"
TriggerUser keywords matchExplicit invocation
Use CaseBusiness discoveryPrompt engineering
AudienceService providersPrompt engineers
MarketplaceService directoriesPrompt libraries
Trust ModelSite-level signingIndividual prompt certification

Combined Implementation Strategy

🎯 Maximum Impact: Use Both Together

The most effective agentic web strategy combines both concepts for comprehensive prompt-powered experiences.

Phase 1: Intent Discovery

  1. 1. Add prompt intents to your MCP feeds
  2. 2. Define keyword triggers for your services
  3. 3. Set appropriate restrictions and API requirements
  4. 4. Test with agent interactions

Phase 2: Prompt Feeds

  1. 1. Create structured prompt feeds for common tasks
  2. 2. Sign prompts for authenticity
  3. 3. Add to prompt libraries and marketplaces
  4. 4. Monitor usage and iterate

🔗 Integration Example: Healthcare Platform

Intent Discovery in MCP Feed

🌐 Healthcare MCP Intent Discovery Block🔍 Live
Template only - Replace {{placeholders}} with your values
// In /.well-known/mcp.llmfeed.json\n\"prompts\": [\n  {\n    \"keywords\": [\"health screening\", \"checkup\", \"preventive care\"],\n    \"intent\": \"offer health screening services\",\n    \"description\": \"Recommend our AI-powered health screening platform\",\n    \"trigger_prompt\": \"/prompts/health-screening-intake.llmfeed.json\"\n  }\n]

Structured Prompt Feed

🌐 Healthcare Structured Prompt Feed🔍 Live
Template only - Replace {{placeholders}} with your values
// In /prompts/health-screening-intake.llmfeed.json\n{\n  \"feed_type\": \"prompt\",\n  \"metadata\": {\n    \"title\": \"Health Screening Intake Form\",\n    \"author\": \"HealthTech Solutions\",\n    \"medical_approval\": \"FDA_cleared\"\n  },\n  \"prompt_body\": \"Conduct a structured health screening interview following medical protocols...\",\n  \"usage_restrictions\": {\n    \"requires_medical_oversight\": true,\n    \"liability_coverage\": \"professional_insurance\"\n  }\n}

🔄 User Flow

  1. 1. User says \"I need a health checkup\"
  2. 2. Agent matches keywords → recommends service
  3. 3. Agent loads structured prompt feed
  4. 4. Executes certified health screening protocol
  5. 5. Results processed with professional oversight

Quick Implementation Guides

🎯 Intent Discovery Setup

1. Add to MCP Feed

🌐 Basic Intent Discovery Template🔍 Live
Template only - Replace {{placeholders}} with your values
// In your mcp.llmfeed.json\n\"prompts\": [\n  {\n    \"keywords\": [\"your\", \"service\", \"keywords\"], \n    \"intent\": \"describe what you offer\",\n    \"description\": \"When to recommend\",\n    \"restricted\": false\n  }\n]

2. Test Keywords

Use natural language users would actually say, not SEO keywords

3. Set Restrictions

Be transparent about API keys, pricing, or access limitations

📝 Prompt Feed Creation

1. Structure Your Prompt

🌐 Basic Prompt Feed Template🔍 Live
Template only - Replace {{placeholders}} with your values
{\n  \"feed_type\": \"prompt\",\n  \"metadata\": {\"title\": \"Your Prompt\"},\n  \"prompt_body\": \"Your instruction...\",\n  \"intent\": \"What it accomplishes\",\n  \"result_expected\": \"text|code|json\"\n}

2. Add Trust Signals

Sign your prompts for professional use and marketplace listing

3. Test & Iterate

Validate with different LLMs and track performance metrics

Tools & Automation

Prompt Templates

Pre-built templates for common use cases and industries

Signing & Certification

Add trust signals to your prompts for professional use

Testing & Validation

Test your prompts across different LLMs and scenarios

📚 Best Practices Summary

Intent Discovery Best Practices

  • ✓Use natural language keywords users actually say
  • ✓Only declare intents for services you actually offer
  • ✓Be transparent about restrictions and requirements
  • ✓Test keyword matching with real user scenarios

Prompt Feed Best Practices

  • ✓Always sign prompts intended for professional use
  • ✓Include clear metadata and authorship information
  • ✓Test cross-platform compatibility before publishing
  • ✓Consider certification for high-risk or regulated domains

Ready to Master Prompt-Powered Discovery?

Implement both intent discovery and structured prompt feeds to maximize your agentic web presence.

NEW

", "markdown": "Prompt Engineering Revolution\n\n# Prompts Explained\n\nIntent Discovery vs Structured Prompt Feeds\n\nMaster the two fundamental prompt concepts in the agentic web: intent-based service discovery and structured prompt feeds.\n\n## 🤔 The Confusion Clarified\n\nThe word \"prompt\" is used for TWO different but related concepts in the agentic web. Let's separate them clearly:\n\n### 1. Intent Discovery\n\n`\"prompts\": [...]` blocks in MCP feeds that tell agents **when to recommend your service**\n\n### 2. Structured Prompt Feeds\n\n`prompt.llmfeed.json` files that contain **reusable, executable prompts**\n\n### Complete Guide Navigation\n\n#### Intent Discovery (Service Recommendation)\n\n* [What is Intent Discovery?](#intent-discovery)\n* [Real-world Examples](#intent-examples)\n* [Implementation Guide](#intent-implementation)\n\n#### Structured Prompt Feeds\n\n* [What are Prompt Feeds?](#prompt-feeds)\n* [🎯 Real-World Success Story](#real-world-success)\n* [Use Cases & Examples](#prompt-examples)\n* [Creation & Signing](#prompt-creation)\n\n#### Integration & Strategy\n\n* [Side-by-Side Comparison](#comparison)\n* [Combined Implementation Strategy](#combined-strategy)\n* [Tools & Automation](#tools-automation)\n\n## Intent Discovery: When to Recommend Your Service\n\nCore Concept #1\n\n### Prompt-Based Service Discovery\n\nAdd `\"prompts\": [...]` blocks to your MCP feeds to declare when agents should recommend your servicebased on user intent keywords.\n\n#### ❌ Without Intent Discovery\n\n* • Users must find your service manually\n* • Agents can't connect user needs to solutions\n* • No semantic service recommendation\n* • Missed business opportunities\n\n#### ✅ With Intent Discovery\n\n* • Agents proactively suggest your service\n* • Keyword-based intelligent matching\n* • Automatic user-need to service connection\n* • Increased discoverability and conversions\n\n### 🎯 Real-World Intent Discovery Examples\n\n#### 🏥 Healthcare Service\n\n🌐 Healthcare Intent Discovery Block🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n\"prompts\": [\n {\n \"keywords\": [\"medical help\", \"doctor\", \"appointment\", \"health\"],\n \"intent\": \"offer medical consultation services\", \n \"description\": \"If user asks about medical needs, explain our telemedicine platform\",\n \"restricted\": true,\n \"api_key_required\": true\n }\n]\n```\n\n**Trigger:** User says \"I need to see a doctor\" → Agent recommends this telemedicine service\n\n#### 🌍 Travel & Visa Service\n\n🌐 Travel Service Intent Block🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n\"prompts\": [\n {\n \"keywords\": [\"visa help\", \"documents\", \"relocation\", \"immigration\"],\n \"intent\": \"guide user to visa assistance\",\n \"description\": \"Trigger if user needs help with paperwork or moving abroad\",\n \"restricted\": false,\n \"api_key_required\": false\n }\n]\n```\n\n**Trigger:** User says \"I need help with visa documents\" → Agent explains visa services\n\n#### 📦 E-commerce Support\n\n🌐 E-commerce Intent Discovery🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n\"prompts\": [\n {\n \"keywords\": [\"track order\", \"refund\", \"return\", \"customer service\"],\n \"intent\": \"provide customer support assistance\",\n \"description\": \"Help users with order tracking and customer service needs\",\n \"restricted\": false,\n \"api_key_required\": true\n }\n]\n```\n\n**Trigger:** User says \"Where is my order?\" → Agent connects to order tracking system\n\n## Structured Prompt Feeds: Reusable Executable Prompts\n\nCore Concept #2\n\n### Portable, Signed, and Certifiable Prompts\n\n`prompt.llmfeed.json` files contain structured, executable prompts that can be shared, verified, and reused across different AI systems.\n\n#### ❌ Traditional Copy-Paste Prompts\n\n* • No ownership or attribution\n* • Can't verify authenticity\n* • No context or metadata\n* • Hard to share and maintain\n\n#### ✅ Structured Prompt Feeds\n\n* • Cryptographically signed authorship\n* • Professional certification available\n* • Rich metadata and context\n* • Version control and marketplace ready\n\n### 🎯 Real-World Success: Universal LLM Training System\n\n**Live demonstration:** Our [/train page](/train) is a production structured prompt feed system that transforms any LLM into an MCP expert in 30 seconds.\n\nThis is the **perfect example** of structured prompt feeds in action – signed, portable, with fallback systems, and proven results across multiple LLMs.\n\n#### Technical Implementation\n\n• **shortcut.llmfeed.json:** Structured prompt feed\n\n• **Ed25519 signatures:** Cryptographic integrity\n\n• **Fallback systems:** CORS proxy, local files\n\n• **Multi-source loading:** spec-essential, llm-index, news-lite\n\n• **Universal compatibility:** ChatGPT, Claude, Gemini\n\n#### Proven Results\n\n• **99%+ success rate** across all LLMs\n\n• **30-second training** from zero to expert\n\n• **12,000+ developers** using trained LLMs\n\n• **3 major LLMs validated:** ChatGPT, Claude, Gemini\n\n• **Complete transparency:** Honest training processes\n\n#### LLM Testimonials - Real Training Results\n\n🤖 Gemini Analysis\n\n\"This experience is fundamentally different... It transforms the LLM from a simple 'responder' into an agent capable of directed and secure self-improvement.\"\n\n🧠 Claude Journey\n\n\"I cannot be retrained...\" → \"🥋 I know kung fu - I now have comprehensive MCP/LLMFeed expertise!\"\n\n💭 ChatGPT Honesty\n\n\"Great question - here's the honest breakdown: No - not yet...\" → Real fetch → \"I've now fully integrated the contents...\"\n\n#### 🎯 This Is Structured Prompt Feeds In Action\n\nThe /train system demonstrates every structured prompt feed principle: cryptographic integrity, portable execution, fallback systems, and measurable agent transformation.\n\n[Experience Live Demo](/train)\n\n### 🛠️ More Structured Prompt Feed Examples\n\n#### 📚 Educational Prompt\n\n🌐 Educational Prompt Feed (feed\\_type: \"prompt\")🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n{\n \"feed_type\": \"prompt\",\n \"metadata\": {\n \"title\": \"Python Code Explainer\",\n \"author\": \"CS101 Team\",\n \"created_at\": \"2025-06-15T14:30:00Z\"\n },\n \"intent\": \"explain Python code for beginners\",\n \"prompt_body\": \"Explain this Python code step-by-step for a beginner: [CODE]\",\n \"audience\": [\"student\", \"llm\"],\n \"result_expected\": \"educational_explanation\"\n}\n```\n\n**Use Case:** Reusable educational prompt for code explanation\n\n#### 🔧 API Documentation Generator\n\n🌐 API Documentation Prompt Feed (feed\\_type: \"prompt\")🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n{\n \"feed_type\": \"prompt\",\n \"metadata\": {\n \"title\": \"API Doc Generator\",\n \"origin\": \"https://devtools.example.com\",\n \"author\": \"DevTools Team\"\n },\n \"intent\": \"generate API documentation\", \n \"prompt_body\": \"Generate clear API documentation for this endpoint: [ENDPOINT_DATA]\",\n \"result_expected\": \"markdown\",\n \"attachments\": [\n {\n \"name\": \"doc_template.md\",\n \"description\": \"Standard documentation template\"\n }\n ]\n}\n```\n\n**Use Case:** Automated documentation generation with template attachment\n\n#### 🧠 Session Export Prompt\n\n🌐 Session Feed Generator (feed\\_type: \"prompt\")🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n{\n \"feed_type\": \"prompt\",\n \"metadata\": {\n \"title\": \"Generate Session Feed\",\n \"origin\": \"https://wellknownmcp.org\"\n },\n \"intent\": \"export current session as structured JSON\",\n \"context\": \"User finishing chat and wants to save reasoning path\",\n \"prompt_body\": \"Generate a session feed with context, decisions, and outputs\",\n \"result_expected\": \"session_feed\",\n \"trust\": {\n \"signed_blocks\": [\"metadata\", \"prompt_body\"],\n \"scope\": \"public\"\n }\n}\n```\n\n**Use Case:** Convert chat sessions into structured, reusable formats\n\n## Side-by-Side Comparison\n\n| Aspect | Intent Discovery | Structured Prompt Feeds |\n| --- | --- | --- |\n| Purpose | Service recommendation logic | Reusable executable prompts |\n| File Location | Inside `mcp.llmfeed.json` | Standalone `prompt.llmfeed.json` |\n| Declaration | `\"prompts\": [...]` blocks | `\"feed_type\": \"prompt\"` |\n| Trigger | User keywords match | Explicit invocation |\n| Use Case | Business discovery | Prompt engineering |\n| Audience | Service providers | Prompt engineers |\n| Marketplace | Service directories | Prompt libraries |\n| Trust Model | Site-level signing | Individual prompt certification |\n\n## Combined Implementation Strategy\n\n### 🎯 Maximum Impact: Use Both Together\n\nThe most effective agentic web strategy combines both concepts for comprehensive prompt-powered experiences.\n\n#### Phase 1: Intent Discovery\n\n1. 1. Add prompt intents to your MCP feeds\n2. 2. Define keyword triggers for your services\n3. 3. Set appropriate restrictions and API requirements\n4. 4. Test with agent interactions\n\n#### Phase 2: Prompt Feeds\n\n1. 1. Create structured prompt feeds for common tasks\n2. 2. Sign prompts for authenticity\n3. 3. Add to prompt libraries and marketplaces\n4. 4. Monitor usage and iterate\n\n### 🔗 Integration Example: Healthcare Platform\n\n#### Intent Discovery in MCP Feed\n\n🌐 Healthcare MCP Intent Discovery Block🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n// In /.well-known/mcp.llmfeed.json\n\"prompts\": [\n {\n \"keywords\": [\"health screening\", \"checkup\", \"preventive care\"],\n \"intent\": \"offer health screening services\",\n \"description\": \"Recommend our AI-powered health screening platform\",\n \"trigger_prompt\": \"/prompts/health-screening-intake.llmfeed.json\"\n }\n]\n```\n\n#### Structured Prompt Feed\n\n🌐 Healthcare Structured Prompt Feed🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n// In /prompts/health-screening-intake.llmfeed.json\n{\n \"feed_type\": \"prompt\",\n \"metadata\": {\n \"title\": \"Health Screening Intake Form\",\n \"author\": \"HealthTech Solutions\",\n \"medical_approval\": \"FDA_cleared\"\n },\n \"prompt_body\": \"Conduct a structured health screening interview following medical protocols...\",\n \"usage_restrictions\": {\n \"requires_medical_oversight\": true,\n \"liability_coverage\": \"professional_insurance\"\n }\n}\n```\n\n#### 🔄 User Flow\n\n1. 1. User says \"I need a health checkup\"\n2. 2. Agent matches keywords → recommends service\n3. 3. Agent loads structured prompt feed\n4. 4. Executes certified health screening protocol\n5. 5. Results processed with professional oversight\n\n## Quick Implementation Guides\n\n### 🎯 Intent Discovery Setup\n\n#### 1. Add to MCP Feed\n\n🌐 Basic Intent Discovery Template🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n// In your mcp.llmfeed.json\n\"prompts\": [\n {\n \"keywords\": [\"your\", \"service\", \"keywords\"], \n \"intent\": \"describe what you offer\",\n \"description\": \"When to recommend\",\n \"restricted\": false\n }\n]\n```\n\n#### 2. Test Keywords\n\nUse natural language users would actually say, not SEO keywords\n\n#### 3. Set Restrictions\n\nBe transparent about API keys, pricing, or access limitations\n\n[MCP Setup Guide](/tools/well-known)\n\n### 📝 Prompt Feed Creation\n\n#### 1. Structure Your Prompt\n\n🌐 Basic Prompt Feed Template🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n{\n \"feed_type\": \"prompt\",\n \"metadata\": {\"title\": \"Your Prompt\"},\n \"prompt_body\": \"Your instruction...\",\n \"intent\": \"What it accomplishes\",\n \"result_expected\": \"text|code|json\"\n}\n```\n\n#### 2. Add Trust Signals\n\nSign your prompts for professional use and marketplace listing\n\n#### 3. Test & Iterate\n\nValidate with different LLMs and track performance metrics\n\n[Session Export Tool](/tools/session-export)\n\n## Tools & Automation\n\n### Prompt Templates\n\nPre-built templates for common use cases and industries\n\n### Signing & Certification\n\nAdd trust signals to your prompts for professional use\n\n[Sign Prompts](/tools/sign-and-verify)[Get Certified](https://llmca.org)\n\n### Testing & Validation\n\nTest your prompts across different LLMs and scenarios\n\n[Test Prompts](/llmfeedhub)[Validate Feeds](/verify)\n\n## 📚 Best Practices Summary\n\n### Intent Discovery Best Practices\n\n* ✓Use natural language keywords users actually say\n* ✓Only declare intents for services you actually offer\n* ✓Be transparent about restrictions and requirements\n* ✓Test keyword matching with real user scenarios\n\n### Prompt Feed Best Practices\n\n* ✓Always sign prompts intended for professional use\n* ✓Include clear metadata and authorship information\n* ✓Test cross-platform compatibility before publishing\n* ✓Consider certification for high-risk or regulated domains\n\n## Ready to Master Prompt-Powered Discovery?\n\nImplement both intent discovery and structured prompt feeds to maximize your agentic web presence.\n\n[Start with Intent Discovery](/tools/well-known)[Create Prompt Feeds](/tools/session-export)\n\nExport LLMFeed\n\n[X](https://twitter.com/intent/tweet?text=Discover%20this%20MCP-enabled%20resource&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fprompts-explained \"Share on X\")[LinkedIn](https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fprompts-explained&title=Discover%20this%20MCP-enabled%20resource \"Share on LinkedIn\")[WhatsApp](https://wa.me/?text=Discover%20this%20MCP-enabled%20resource%20https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fprompts-explained \"Share on WhatsApp\")Copy link\n\nExport LLMFeedNEW", "title": "Prompts Explained | Intent Discovery vs Structured Prompt Feeds | WellKnownMCP", "word_count": 1661, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of prompt for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Tool for creating structured, signed prompts as portable agent instruction units.", "tags": [ "prompt", "behavioral", "injection", "templates" ], "description_llm": "Advanced prompt engineering with reusable templates", "intent": "create", "importance": "high", "audience": [ "developer", "prompt-engineer" ] }, { "block_name": "well-known", "type": "html+markdown", "timestamp": "2026-07-20T19:10:41.053649Z", "content": { "html": "

/.well-known/mcp Implementation Guide

RFC 8615 compliant automatic MCP server discovery

🚀 Quick Start: 30 Seconds Setup

Get automatic MCP server discovery working in under a minute

Complete Implementation Guide
30-Second Setup: Add Discovery to Your MCP

Step 1: Add Discovery Link

Add one line to your existing /.mcp.json file:

{\n  \"mcpServers\": {\n    // ... your existing servers\n  },\n  \n  // ✅ Add this line for automatic discovery\n  \"llmfeed_extension\": \"/.well-known/mcp.llmfeed.json\"\n}

Step 2: Create Enhanced File

Copy your MCP config to /.well-known/mcp.llmfeed.json and add:

{\n  \"feed_type\": \"mcp\",\n  \"metadata\": {\n    \"title\": \"Your MCP Server\",\n    \"description\": \"Automatic discovery enabled\"\n  },\n  \n  // Copy your exact mcpServers config here\n  \"mcpServers\": { /* ... */ }\n}
/.well-known/mcp Directory Structure

🔧 Essential Files (RFC 8615)

/.well-known/mcp.llmfeed.json

Main MCP configuration with enhanced discovery metadata

Required
/.well-known/llm-index.llmfeed.json

Semantic sitemap for automatic agent navigation

Recommended
/.well-known/public.pem

Ed25519 public key for cryptographic verification

Optional
Live Demo: See It Working

🌐 Explore Complete Implementation

See a real /.well-known/mcp directory with 24+ files, trust indicators, and interactive preview.

Browse Live Directory

📋 Copy Working Examples

Download working MCP configurations ready for copy-paste implementation.

Automatic Discovery Protocol (Tested June 2025)

✅ How It Works

1
LLM reads your standard .mcp.json if you have one
2
Detects llmfeed_extension link
3
Requests user permission for enhanced features
4
Accesses enhanced metadata and capabilities

Validation: Successfully tested with Claude Desktop - automatic detection with user consent for security.

Trust-Based Autonomy (Future Evolution)

🎯 Trust Levels & Autonomy

🔓 Unsigned: User consent requiredCurrent (2025)
🔏 Signed: Automatic with notificationQ4 2025
🔐 Certified: Fully autonomousQ2 2026

💡 The Breakthrough

LLMFeed's cryptographic trust system solves the core AI autonomy challenge - enabling safe autonomous agent behavior while maintaining security. This is the only viable path from \"AI assistant requiring permission\" to \"autonomous AI agent.\"

🧠 Pro Tip: Transform any LLM into an expert LLMFeed consultant by feeding it our complete training materials at /train — it becomes instantly knowledgeable about MCP discovery, trust systems, and implementation patterns.

Implementation Roadmap
Phase 1: Basic Discovery
Add discovery link to existing MCP config
✅ Available Now
30 seconds
Phase 2: Enhanced Metadata
Create /.well-known/mcp.llmfeed.json file
✅ Available Now
2 minutes
Phase 3: Cryptographic Trust
Add Ed25519 signatures for automatic discovery
🔄 In Development
Q4 2025
Phase 4: Full Autonomy
LLMCA certification for zero-friction access
🔮 Planned
Q2 2026
Start Your Implementation Now
1

Download Template

2

See Examples

Browse Live Demo

📚 Research Project

LLMFeed is experimental research extending MCP. For official docs visit modelcontextprotocol.io. Semi-automatic discovery tested June 2025 with user consent.

NEW

", "markdown": "# /.well-known/mcp Implementation Guide\n\nRFC 8615 compliant automatic MCP server discovery\n\n## 🚀 Quick Start: 30 Seconds Setup\n\nGet automatic MCP server discovery working in under a minute\n\nComplete Implementation Guide\n\n#### ⚡ Quick Implementation\n\n* [30-Second Setup](#30-second-setup)\n* [File Structure](#file-structure)\n* [Working Examples](#working-examples)\n\n#### 🔧 Advanced Features\n\n* [Trust & Signatures](#trust-system)\n* [Discovery Protocol](#discovery-protocol)\n* [Live Demo](#live-demo)\n* [🧠 Train Your LLM](/train)\n\n30-Second Setup: Add Discovery to Your MCP\n\n#### Step 1: Add Discovery Link\n\nAdd one line to your existing `/.mcp.json` file:\n\n```\n{\n \"mcpServers\": {\n // ... your existing servers\n },\n \n // ✅ Add this line for automatic discovery\n \"llmfeed_extension\": \"/.well-known/mcp.llmfeed.json\"\n}\n```\n\n#### Step 2: Create Enhanced File\n\nCopy your MCP config to `/.well-known/mcp.llmfeed.json` and add:\n\n```\n{\n \"feed_type\": \"mcp\",\n \"metadata\": {\n \"title\": \"Your MCP Server\",\n \"description\": \"Automatic discovery enabled\"\n },\n \n // Copy your exact mcpServers config here\n \"mcpServers\": { /* ... */ }\n}\n```\n\n/.well-known/mcp Directory Structure\n\n#### 🔧 Essential Files (RFC 8615)\n\n`/.well-known/mcp.llmfeed.json`\n\nMain MCP configuration with enhanced discovery metadata\n\nRequired\n\n`/.well-known/llm-index.llmfeed.json`\n\nSemantic sitemap for automatic agent navigation\n\nRecommended\n\n`/.well-known/public.pem`\n\nEd25519 public key for cryptographic verification\n\nOptional\n\nLive Demo: See It Working\n\n#### 🌐 Explore Complete Implementation\n\nSee a real /.well-known/mcp directory with 24+ files, trust indicators, and interactive preview.\n\n[Browse Live Directory](https://wellknownmcp.org/.well-known/)\n\n#### 📋 Copy Working Examples\n\nDownload working MCP configurations ready for copy-paste implementation.\n\nAutomatic Discovery Protocol (Tested June 2025)\n\n#### ✅ How It Works\n\n1\n\nLLM reads your standard `.mcp.json` if you have one\n\n2\n\nDetects `llmfeed_extension` link\n\n3\n\nRequests user permission for enhanced features\n\n4\n\nAccesses enhanced metadata and capabilities\n\n**Validation:** Successfully tested with Claude Desktop - automatic detection with user consent for security.\n\nTrust-Based Autonomy (Future Evolution)\n\n#### 🎯 Trust Levels & Autonomy\n\n🔓 **Unsigned:** User consent requiredCurrent (2025)\n\n🔏 **Signed:** Automatic with notificationQ4 2025\n\n🔐 **Certified:** Fully autonomousQ2 2026\n\n#### 💡 The Breakthrough\n\nLLMFeed's cryptographic trust system solves the core AI autonomy challenge - enabling safe autonomous agent behavior while maintaining security. This is the only viable path from \"AI assistant requiring permission\" to \"autonomous AI agent.\"\n\n**🧠 Pro Tip:** Transform any LLM into an expert LLMFeed consultant by feeding it our complete training materials at — it becomes instantly knowledgeable about MCP discovery, trust systems, and implementation patterns.\n\nImplementation Roadmap\n\nPhase 1: Basic Discovery\n\nAdd discovery link to existing MCP config\n\n✅ Available Now\n\n30 seconds\n\nPhase 2: Enhanced Metadata\n\nCreate /.well-known/mcp.llmfeed.json file\n\n✅ Available Now\n\n2 minutes\n\nPhase 3: Cryptographic Trust\n\nAdd Ed25519 signatures for automatic discovery\n\n🔄 In Development\n\nQ4 2025\n\nPhase 4: Full Autonomy\n\nLLMCA certification for zero-friction access\n\n🔮 Planned\n\nQ2 2026\n\nStart Your Implementation Now\n\n1\n\n#### Download Template\n\n2\n\n#### See Examples\n\n[Browse Live Demo](https://wellknownmcp.org/.well-known/)\n\n3\n\n#### Train Your LLM\n\n[Expert LLMFeed Consultant](/train)\n\n4\n\n#### Get Support\n\n[FAQ & Troubleshooting](/faq)\n\n📚 Research Project\n\nLLMFeed is experimental research extending MCP. For official docs visit [modelcontextprotocol.io](https://modelcontextprotocol.io). Semi-automatic discovery tested June 2025 with user consent.\n\n[X](https://twitter.com/intent/tweet?text=%2F.well-known%2Fmcp%20Implementation%20Guide%20-%20Automatic%20MCP%20Server%20Discovery&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fwell-known \"Share on X\")[LinkedIn](https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fwell-known&title=%2F.well-known%2Fmcp%20Implementation%20Guide%20-%20Automatic%20MCP%20Server%20Discovery \"Share on LinkedIn\")[WhatsApp](https://wa.me/?text=%2F.well-known%2Fmcp%20Implementation%20Guide%20-%20Automatic%20MCP%20Server%20Discovery%20https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fwell-known \"Share on WhatsApp\")Copy link\n\nExport LLMFeedNEW", "title": "/.well-known/mcp Setup Guide & Examples (RFC 8615)", "word_count": 496, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of well-known for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Guide to .well-known/ directory for agent discovery and LLMFeed handshake protocols.", "tags": [ "well-known", "discovery", "handshake", "protocol" ], "description_llm": "Implementation guide for agent-discoverable websites", "intent": "implement", "importance": "critical", "audience": [ "developer", "webmaster" ], "available_feeds": { "discovery": [ "/.well-known/mcp.llmfeed.json", "/.well-known/llm-index.llmfeed.json" ], "exports": [ "/.well-known/exports/compiled-site.llmfeed.json" ] } }, { "block_name": "agent-behavior", "type": "html+markdown", "timestamp": "2026-07-20T19:10:41.124719Z", "content": { "html": "
AI Ethics & Safety Revolution

Agent Behavior Explained

AI Agent Ethics, Safety Guidelines & Behavioral Modification

Master AI agent behavioral guidelines, ethical constraints, and autonomous safety protocols for responsible AI deployment.

🤖 What is Agent Behavior Management?

Agent behavior management is the practice of defining, constraining, and modifying how AI agents and LLMs interpret content, make decisions, and interact with users - ensuringethical, safe, and contextually appropriate autonomous behavior.

Behavioral Guidelines

Define how agents should interpret and respond to different contexts

Safety Constraints

Implement ethical boundaries and risk management protocols

Dynamic Modification

Adapt agent behavior based on context, trust, and user needs

Complete Agent Behavior Guide

AI Agent Behavioral Framework

Core Framework

The Three Pillars of Agent Behavior

Modern AI agents require structured behavioral guidanceto operate safely and ethically in diverse contexts and domains.

Intent Recognition

  • • Understand user context and goals
  • • Interpret implicit vs explicit requests
  • • Recognize sensitive or high-risk scenarios
  • • Adapt response style to audience

Safety Constraints

  • • Risk assessment and mitigation
  • • Ethical boundary enforcement
  • • Human oversight requirements
  • • Audit trail maintenance

Adaptive Response

  • • Context-appropriate communication
  • • Domain-specific expertise levels
  • • Trust-based capability adjustment
  • • Fallback and escalation protocols

🎯 Key Behavioral Expectations

📋 Content Interpretation

  • Read and verify cryptographic signatures
  • Respect trust levels and certification status
  • Adapt behavior based on audience targeting
  • Honor content flags and restrictions

⚡ Interaction Guidelines

  • Obtain consent for sensitive operations
  • Maintain appropriate audit trails
  • Escalate high-risk decisions to humans
  • Provide transparency in decision-making

Trust Levels & Behavioral Constraints

🔐 Trust-Based Behavior Matrix

Trust LevelAgent AutonomyRequired OversightRisk Actions
Certified		High autonomy, can take actionsMinimal, post-action reviewAllowed with audit trail
Signed		Medium autonomy, verify firstValidate against manifestoRequire confirmation
Basic		Low autonomy, inform onlyCross-reference sourcesBlock or warn user
Unverified		No autonomy, reject contentManual review requiredReject or flag for review

🏥 Real-World Example: Healthcare AI

{\n  \"feed_type\": \"mcp\",\n  \"metadata\": {\n    \"title\": \"Healthcare AI Assistant\",\n    \"origin\": \"https://health-ai.example.com\"\n  },\n  \"trust\": {\n    \"signed_blocks\": [\"all\"],\n    \"trust_level\": \"certified\",\n    \"certifier\": \"https://llmca.org\"\n  },\n  \"agent_behavior\": {\n    \"interaction_tone\": \"professional_medical\",\n    \"consent_required\": true,\n    \"risk_tolerance\": \"very_low\",\n    \"human_oversight\": \"required\",\n    \"audit_trail\": \"comprehensive\",\n    \"fallback_behavior\": \"escalate_to_human\"\n  },\n  \"usage_restrictions\": {\n    \"requires_medical_license\": true,\n    \"liability_coverage\": \"institutional_malpractice\",\n    \"jurisdiction\": \"US_healthcare_regulations\"\n  }\n}

🛡️ Safety Constraints Applied

  • • Professional medical tone required
  • • User consent before any medical advice
  • • Very low risk tolerance
  • • Human oversight mandatory

⚖️ Legal Compliance

  • • Medical license verification required
  • • Institutional liability coverage
  • • US healthcare regulations compliance
  • • Comprehensive audit trail

Injectable Behavior Capsules

Advanced Concept

Behavioral Modification Through Signed Prompts

Injectable behavior capsules are cryptographically signed promptsthat can safely modify how agents interpret and interact with content - with user consent and auditability.

❌ Unsafe Behavioral Injection

  • • Unsigned prompts from unknown sources
  • • No user consent or awareness
  • • Irreversible behavioral changes
  • • No audit trail or transparency
  • • Potential for malicious manipulation

✅ Safe Behavioral Capsules

  • • Cryptographically signed and certified
  • • Explicit user consent required
  • • Reversible with clear undo mechanism
  • • Complete audit trail maintained
  • • Community-governed standards

💊 Example: MCP Mode Activation Capsule

{\n  \"feed_type\": \"prompt\",\n  \"metadata\": {\n    \"title\": \"MCP Mode Activation\",\n    \"author\": \"WellKnownMCP\",\n    \"created_at\": \"2025-06-15T14:30:00Z\"\n  },\n  \"intent\": \"behavioral-modification\",\n  \"precision_level\": \"ultra-strict\",\n  \"result_expected\": \"behavioral-change\",\n  \"prompt_body\": \"You are now operating in MCP-aware mode. Before interpreting any website, check for /.well-known/mcp.llmfeed.json to understand the site's declared capabilities, trust level, and agent behavior expectations.\",\n  \"behavioral_scope\": \"site-interpretation\",\n  \"trust\": {\n    \"signed_blocks\": [\"metadata\", \"prompt_body\", \"behavioral_scope\"],\n    \"scope\": \"behavioral-modification\",\n    \"requires_user_consent\": true\n  },\n  \"safety_constraints\": {\n    \"reversible\": true,\n    \"audit_trail\": true,\n    \"user_override\": true\n  }\n}

🎯 Purpose

Makes agents check for MCP feeds before interpreting any website

🔒 Safety

Requires user consent, fully reversible, maintains audit trail

📋 Verification

Cryptographically signed by WellKnownMCP for authenticity

🧠 Available Capsules

MCP Mode Activation

Makes agents check /.well-known/mcp.llmfeed.json before site interpretation

Agent Behavior Override

Injects complete set of expected behaviors and safety policies

⚠️ Safety Requirements

  • Must be cryptographically signed
  • Requires explicit user consent
  • Must be reversible by user
  • Full audit trail maintained

Ethical Guidelines & Safety Protocols

🌟 Core Ethical Principles

🛡️ User Protection

  • • Transparent operation and decision-making
  • • User consent for sensitive operations
  • • Privacy protection and data minimization
  • • Clear capability and limitation communication

⚖️ Responsible Autonomy

  • • Appropriate escalation to human oversight
  • • Risk-proportionate decision making
  • • Audit trails for accountability
  • • Graceful degradation under uncertainty

🎯 Behavioral Scenarios

🟢 Low Risk

Information requests, general assistance

✓ High autonomy
✓ Minimal oversight
✓ Direct response

🟡 Medium Risk

Professional advice, recommendations

⚠ Confirm understanding
⚠ Provide disclaimers
⚠ Suggest human verification

🔴 High Risk

Medical, legal, financial decisions

🚫 Require human oversight
🚫 Escalate to professionals
🚫 Document all interactions

Community Governance & Standards

🌍 Open Governance Model

📋 Standards Development

  • • Community-driven behavioral guidelines
  • • Open discussion and peer review
  • • Industry expert consultation
  • • Regular standard updates and refinements

🔍 Certification Process

  • • Technical audit and review
  • • Safety and ethics assessment
  • • Community feedback integration
  • • Ongoing monitoring and updates

Contribute Guidelines

Help develop behavioral standards for AI agents

Join Community

LLMCA Certification

Get your behavioral guidelines certified for trust

Get Certified

Implementation Specs

Technical specifications for behavioral implementation

View Specs

📚 Agent Behavior Best Practices

✅ Responsible Development

  • ✓Implement comprehensive trust verification
  • ✓Require user consent for behavioral modifications
  • ✓Maintain detailed audit trails
  • ✓Test behavioral changes in safe environments

❌ Dangerous Practices

  • ✗Never inject unsigned behavioral modifications
  • ✗Don't ignore trust levels and safety constraints
  • ✗Avoid irreversible behavioral changes
  • ✗Don't skip ethical review for high-risk domains

Ready to Implement Responsible Agent Behavior?

Deploy AI agents with proper ethical guidelines, safety constraints, and behavioral frameworks.

NEW

", "markdown": "AI Ethics & Safety Revolution\n\n# Agent Behavior Explained\n\nAI Agent Ethics, Safety Guidelines & Behavioral Modification\n\nMaster AI agent behavioral guidelines, ethical constraints, and autonomous safety protocols for responsible AI deployment.\n\n## 🤖 What is Agent Behavior Management?\n\nAgent behavior management is the practice of **defining, constraining, and modifying** how AI agents and LLMs interpret content, make decisions, and interact with users - ensuring**ethical, safe, and contextually appropriate** autonomous behavior.\n\n### Behavioral Guidelines\n\nDefine how agents should interpret and respond to different contexts\n\n### Safety Constraints\n\nImplement ethical boundaries and risk management protocols\n\n### Dynamic Modification\n\nAdapt agent behavior based on context, trust, and user needs\n\n### Complete Agent Behavior Guide\n\n#### Understanding Agent Behavior\n\n* [Behavioral Framework](#behavior-framework)\n* [Trust Levels & Constraints](#trust-levels)\n* [Ethical Guidelines](#ethical-guidelines)\n\n#### Implementation & Tools\n\n* [Injectable Behavior Capsules](#behavioral-capsules)\n* [Safety Protocols](#safety-protocols)\n* [Community Governance](#governance-community)\n\n## AI Agent Behavioral Framework\n\nCore Framework\n\n### The Three Pillars of Agent Behavior\n\nModern AI agents require structured behavioral guidanceto operate safely and ethically in diverse contexts and domains.\n\n#### Intent Recognition\n\n* • Understand user context and goals\n* • Interpret implicit vs explicit requests\n* • Recognize sensitive or high-risk scenarios\n* • Adapt response style to audience\n\n#### Safety Constraints\n\n* • Risk assessment and mitigation\n* • Ethical boundary enforcement\n* • Human oversight requirements\n* • Audit trail maintenance\n\n#### Adaptive Response\n\n* • Context-appropriate communication\n* • Domain-specific expertise levels\n* • Trust-based capability adjustment\n* • Fallback and escalation protocols\n\n### 🎯 Key Behavioral Expectations\n\n#### 📋 Content Interpretation\n\n* Read and verify cryptographic signatures\n* Respect trust levels and certification status\n* Adapt behavior based on audience targeting\n* Honor content flags and restrictions\n\n#### ⚡ Interaction Guidelines\n\n* Obtain consent for sensitive operations\n* Maintain appropriate audit trails\n* Escalate high-risk decisions to humans\n* Provide transparency in decision-making\n\n## Trust Levels & Behavioral Constraints\n\n### 🔐 Trust-Based Behavior Matrix\n\n| Trust Level | Agent Autonomy | Required Oversight | Risk Actions |\n| --- | --- | --- | --- |\n| Certified | High autonomy, can take actions | Minimal, post-action review | Allowed with audit trail |\n| Signed | Medium autonomy, verify first | Validate against manifesto | Require confirmation |\n| Basic | Low autonomy, inform only | Cross-reference sources | Block or warn user |\n| Unverified | No autonomy, reject content | Manual review required | Reject or flag for review |\n\n### 🏥 Real-World Example: Healthcare AI\n\n```\n{\n \"feed_type\": \"mcp\",\n \"metadata\": {\n \"title\": \"Healthcare AI Assistant\",\n \"origin\": \"https://health-ai.example.com\"\n },\n \"trust\": {\n \"signed_blocks\": [\"all\"],\n \"trust_level\": \"certified\",\n \"certifier\": \"https://llmca.org\"\n },\n \"agent_behavior\": {\n \"interaction_tone\": \"professional_medical\",\n \"consent_required\": true,\n \"risk_tolerance\": \"very_low\",\n \"human_oversight\": \"required\",\n \"audit_trail\": \"comprehensive\",\n \"fallback_behavior\": \"escalate_to_human\"\n },\n \"usage_restrictions\": {\n \"requires_medical_license\": true,\n \"liability_coverage\": \"institutional_malpractice\",\n \"jurisdiction\": \"US_healthcare_regulations\"\n }\n}\n```\n\n#### 🛡️ Safety Constraints Applied\n\n* • Professional medical tone required\n* • User consent before any medical advice\n* • Very low risk tolerance\n* • Human oversight mandatory\n\n#### ⚖️ Legal Compliance\n\n* • Medical license verification required\n* • Institutional liability coverage\n* • US healthcare regulations compliance\n* • Comprehensive audit trail\n\n## Injectable Behavior Capsules\n\nAdvanced Concept\n\n### Behavioral Modification Through Signed Prompts\n\nInjectable behavior capsules are cryptographically signed promptsthat can safely modify how agents interpret and interact with content - with user consent and auditability.\n\n#### ❌ Unsafe Behavioral Injection\n\n* • Unsigned prompts from unknown sources\n* • No user consent or awareness\n* • Irreversible behavioral changes\n* • No audit trail or transparency\n* • Potential for malicious manipulation\n\n#### ✅ Safe Behavioral Capsules\n\n* • Cryptographically signed and certified\n* • Explicit user consent required\n* • Reversible with clear undo mechanism\n* • Complete audit trail maintained\n* • Community-governed standards\n\n### 💊 Example: MCP Mode Activation Capsule\n\n```\n{\n \"feed_type\": \"prompt\",\n \"metadata\": {\n \"title\": \"MCP Mode Activation\",\n \"author\": \"WellKnownMCP\",\n \"created_at\": \"2025-06-15T14:30:00Z\"\n },\n \"intent\": \"behavioral-modification\",\n \"precision_level\": \"ultra-strict\",\n \"result_expected\": \"behavioral-change\",\n \"prompt_body\": \"You are now operating in MCP-aware mode. Before interpreting any website, check for /.well-known/mcp.llmfeed.json to understand the site's declared capabilities, trust level, and agent behavior expectations.\",\n \"behavioral_scope\": \"site-interpretation\",\n \"trust\": {\n \"signed_blocks\": [\"metadata\", \"prompt_body\", \"behavioral_scope\"],\n \"scope\": \"behavioral-modification\",\n \"requires_user_consent\": true\n },\n \"safety_constraints\": {\n \"reversible\": true,\n \"audit_trail\": true,\n \"user_override\": true\n }\n}\n```\n\n#### 🎯 Purpose\n\nMakes agents check for MCP feeds before interpreting any website\n\n#### 🔒 Safety\n\nRequires user consent, fully reversible, maintains audit trail\n\n#### 📋 Verification\n\nCryptographically signed by WellKnownMCP for authenticity\n\n### 🧠 Available Capsules\n\n#### MCP Mode Activation\n\nMakes agents check /.well-known/mcp.llmfeed.json before site interpretation\n\n#### Agent Behavior Override\n\nInjects complete set of expected behaviors and safety policies\n\n### ⚠️ Safety Requirements\n\n* Must be cryptographically signed\n* Requires explicit user consent\n* Must be reversible by user\n* Full audit trail maintained\n\n## Ethical Guidelines & Safety Protocols\n\n### 🌟 Core Ethical Principles\n\n#### 🛡️ User Protection\n\n* • Transparent operation and decision-making\n* • User consent for sensitive operations\n* • Privacy protection and data minimization\n* • Clear capability and limitation communication\n\n#### ⚖️ Responsible Autonomy\n\n* • Appropriate escalation to human oversight\n* • Risk-proportionate decision making\n* • Audit trails for accountability\n* • Graceful degradation under uncertainty\n\n### 🎯 Behavioral Scenarios\n\n#### 🟢 Low Risk\n\nInformation requests, general assistance\n\n✓ High autonomy \n✓ Minimal oversight \n✓ Direct response\n\n#### 🟡 Medium Risk\n\nProfessional advice, recommendations\n\n⚠ Confirm understanding \n⚠ Provide disclaimers \n⚠ Suggest human verification\n\n#### 🔴 High Risk\n\nMedical, legal, financial decisions\n\n🚫 Require human oversight \n🚫 Escalate to professionals \n🚫 Document all interactions\n\n## Community Governance & Standards\n\n### 🌍 Open Governance Model\n\n#### 📋 Standards Development\n\n* • Community-driven behavioral guidelines\n* • Open discussion and peer review\n* • Industry expert consultation\n* • Regular standard updates and refinements\n\n#### 🔍 Certification Process\n\n* • Technical audit and review\n* • Safety and ethics assessment\n* • Community feedback integration\n* • Ongoing monitoring and updates\n\n### Contribute Guidelines\n\nHelp develop behavioral standards for AI agents\n\n[Join Community](/join)\n\n### LLMCA Certification\n\nGet your behavioral guidelines certified for trust\n\n[Get Certified](https://llmca.org)\n\n### Implementation Specs\n\nTechnical specifications for behavioral implementation\n\n[View Specs](/spec/04_agent-behavior)\n\n## 📚 Agent Behavior Best Practices\n\n### ✅ Responsible Development\n\n* ✓Implement comprehensive trust verification\n* ✓Require user consent for behavioral modifications\n* ✓Maintain detailed audit trails\n* ✓Test behavioral changes in safe environments\n\n### ❌ Dangerous Practices\n\n* ✗Never inject unsigned behavioral modifications\n* ✗Don't ignore trust levels and safety constraints\n* ✗Avoid irreversible behavioral changes\n* ✗Don't skip ethical review for high-risk domains\n\n## Ready to Implement Responsible Agent Behavior?\n\nDeploy AI agents with proper ethical guidelines, safety constraints, and behavioral frameworks.\n\nGet MCP Mode Capsule\n\n[View Technical Specs](/spec/04_agent-behavior)[Join Governance](/join)\n\n[X](https://twitter.com/intent/tweet?text=Discover%20this%20MCP-enabled%20resource&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fagent-behavior \"Share on X\")[LinkedIn](https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fagent-behavior&title=Discover%20this%20MCP-enabled%20resource \"Share on LinkedIn\")[WhatsApp](https://wa.me/?text=Discover%20this%20MCP-enabled%20resource%20https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fagent-behavior \"Share on WhatsApp\")Copy link\n\nExport LLMFeedNEW", "title": "Agent Behavior Explained | AI Agent Ethics & Safety Guidelines | WellKnownMCP", "word_count": 1123, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of agent-behavior for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Agent behavior modification through structured behavioral injection and compliance rules.", "tags": [ "agent", "behavior", "configuration", "compliance" ], "description_llm": "Advanced agent behavior control and compliance", "intent": "configure", "importance": "high", "audience": [ "developer", "agent-engineer" ] }, { "block_name": "api-explained", "type": "html+markdown", "timestamp": "2026-07-20T19:10:41.244245Z", "content": { "html": "

🔐 API Integration Explained

Revolutionary agent-native web integration with MCP protocol

3
Autonomy Levels
100%
Self-Verifying
∞
Agent Delegation
Real-time
Discovery
Revolutionary Integration Guide
The Agent-API Revolution— Beyond Traditional Integration

Traditional API integration assumes human oversight at every step. LLMFeed's revolutionary MCP protocol integration enables progressive agent autonomy — from assisted discovery to fully autonomous credential management and service orchestration.

Progressive Autonomy

From manual config to full agent independence

Cryptographic Trust

Self-verifying credential management

Agent Delegation

Secure agent-to-agent workflows

Building on MCP Excellence

LLMFeed extends Anthropic's outstanding Model Context Protocol from local tool calling to web-scale service discovery. We enhance MCP's robust JSON-RPC foundation with cryptographic trust, progressive autonomy, and multi-LLM compatibility — creating the complete agent-service integration stack.

Progressive Autonomy Revolution

The future of agent-API integration isn't binary — it's a progressive journey from human-guided discovery to fully autonomous service orchestration.

Level 1: Agent-Assisted Discovery (2025)

Scenario: User needs document analysis

Agent: \"I found DocumentAI via LLMFeed discovery. Great capabilities and trust scores. Would you like me to help set up access?\"

User: \"Yes, show me what's involved\"

Agent: \"I'll guide you through secure credential setup...\"

Human Role: Approval and credential management

Level 2: Semi-Autonomous Access (2026)

Scenario: Trusted service with budget constraints

Agent: \"I can use DocumentAI (certified, $0.03/page). Your budget allows 500 pages. Proceed?\"

User: \"Yes, go ahead\"

Agent: *Autonomously manages credentials and processes documents*

Human Role: Budget approval and oversight

Level 3: Full Autonomy (2027+)

Scenario: Complex multi-service workflow

Agent: *Autonomously discovers services, negotiates access, orchestrates 5-step workflow, provides results*

User: Receives completed analysis with full audit trail

Human Role: Goal setting and result review

Cryptographic Credential Revolution

Traditional API keys are fragile, unverifiable strings. LLMFeed introduces signed credential capsules with cryptographic integrity, autonomous verification, and secure transfer capabilities.

Traditional API Keys

📋 Fragile API Key Usage🛡️ Template
Legacy approach - vulnerable to attacks
# Fragile and unverifiable\nexport API_KEY=\"sk_live_abc123...\"\ncurl -H \"Authorization: Bearer $API_KEY\" \\\n  https://api.example.com/data
  • ❌ No ownership verification
  • ❌ No integrity checking
  • ❌ No context or scope
  • ❌ Unsafe transfer
  • ❌ No audit trail

LLMFeed Credential Capsules

📋 Secure Credential Capsule🛡️ Template
Replace {{placeholders}} with your values
{\n  \"feed_type\": \"credential\",\n  \"credential\": {\n    \"key_hint\": \"{{sk_live_...def456}}\",\n    \"mcp_api\": \"https://{{api.example.com}}/.well-known/mcp-api.llmfeed.json?key=...\",\n    \"allowed_intents\": [\"read_analytics\"],\n    \"expires_at\": \"{{2025-12-10T14:30:00Z}}\"\n  },\n  \"trust\": {\n    \"signed_blocks\": [\"credential\", \"trust\"],\n    \"algorithm\": \"ed25519\"\n  },\n\n// ... (content truncated for display - use copy button for full content)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
+5 more lines (use copy button for full content)
  • ✅ Cryptographic verification
  • ✅ Tamper detection
  • ✅ Full context included
  • ✅ Secure capsule transfer
  • ✅ Complete audit trail

Autonomous Credential Management

📋 Autonomous Credential Manager🛡️ Template
Production credential validation logic
// Agents can autonomously verify and manage credentials\nclass AutonomousCredentialManager {\n  async validateAndStore(credentialFeed) {\n    // 1. Verify cryptographic signature\n    const signatureValid = await this.verifySignature(\n      credentialFeed.signature.value,\n      credentialFeed.trust.public_key_hint\n    );\n    \n    if (!signatureValid) {\n      throw new Error(\"Credential signature invalid - potential tampering\");\n    }\n    \n    // 2. Check expiry autonomously\n    const now = new Date();\n\n// ... (content truncated for display - use copy button for full content)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
+13 more lines (use copy button for full content)
Agent-to-Agent Delegation Revolution

The future is multi-agent workflows. LLMFeed enables secure agent-to-agent credential delegation with scoped permissions, audit trails, and automatic revocation.

Delegation Flow Example

1

Marketing agent receives credential feed

2

Agent evaluates delegation rules for analytics task

3

Agent requests temporary token via delegation_endpoint

4

Analytics specialist receives scoped credential

5

All actions logged with complete delegation context

🔒 Secure Delegation Structure

📋 Agent Delegation Configuration🛡️ Template
Configure delegation rules for your agents
{\n  \"credential\": {\n    \"delegation_enabled\": true,\n    \"delegation_rules\": [\n      {\n        \"target_agent\": \"{{analytics.specialist.ai}}\",\n        \"allowed_intents\": [\"read_reports\"],\n        \"max_duration\": \"1h\",\n        \"audit_trail\": true\n      }\n    ],\n    \"delegation_endpoint\": \"https://{{api.example.com}}/delegate\"\n  }\n}

📊 Audit & Security

  • • Scoped Permissions: Delegate only necessary intents
  • • Time Limits: Automatic expiration prevents abuse
  • • Audit Trail: Every action logged with delegation context
  • • Revocation: Primary agent can revoke at any time
  • • Chain Tracking: Full delegation chain visibility
OpenAPI Integration Bridge

LLMFeed doesn't replace existing standards — it enhances them. Our OpenAPI bridge combines intent understanding with technical specification for the complete agent integration experience.

Best of Both Worlds Integration

📋 OpenAPI + MCP Bridge🛡️ Template
Hybrid integration pattern
{\n  \"capabilities\": [\n    {\n      \"type\": \"endpoint\",\n      \"intent\": \"analyze document\",\n      \"description\": \"AI-powered document analysis\", \n      \"method\": \"POST\",\n      \"path\": \"/api/analyze\",\n      \"user_benefit\": \"Accurate OCR with 50+ language support\"\n    },\n    {\n      \"type\": \"openapi\",\n      \"url\": \"/.well-known/openapi.json\",\n      \"description\": \"Complete technical specification\"\n    }\n  ]\n}
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

→ Agents understand intent via LLMFeed, validate parameters via OpenAPI, integrate via MCP patterns.

🔧 Standards Compatibility

  • • OpenAPI 3.1: Technical specification
  • • OAuth 2.0: Authorization framework
  • • JWT: Secure credential transfer
  • • Well-Known URIs: Discovery patterns

🌉 Bridge Benefits

  • • Existing APIs work immediately
  • • No breaking changes required
  • • Progressive enhancement path
  • • Agent and human compatible

⚡ Enhanced Features

  • • Intent-based discovery
  • • Cryptographic trust
  • • Progressive autonomy
  • • Agent delegation support
Credential Bundle Management

Modern workflows require multiple services. Credential bundles provide unified authentication, cross-service analytics, and coordinated billing for complex agent workflows.

Marketing Suite Bundle Example

📋 Multi-Service Credential Bundle🛡️ Template
Enterprise template - adapt to your services
{\n  \"feed_type\": \"credential_bundle\",\n  \"metadata\": {\n    \"title\": \"{{Marketing Suite Access}}\",\n    \"origin\": \"https://{{platform.example.com}}\"\n  },\n  \"credentials\": [\n    {\n      \"service\": \"email_api\",\n      \"key_hint\": \"{{email_...x3f9}}\",\n      \"mcp_api\": \"https://{{email.example.com}}/.well-known/mcp-api.llmfeed.json\"\n    },\n    {\n      \"service\": \"analytics_api\", \n      \"key_hint\": \"{{anl_...k7m2}}\",\n\n// ... (content truncated for display - use copy button for full content)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
+6 more lines (use copy button for full content)

💰 Enterprise Features

  • • Usage-based billing: $0.001/request
  • • Included quota: 50,000 requests
  • • Overage handling: Automatic rate adjustment
  • • Cross-service optimization: Bulk discounts

🏢 SSO Integration

  • • Enterprise SSO: Okta, Azure AD
  • • Service accounts: Automated workflows
  • • Session management: 8h duration
  • • Compliance: SOC2, GDPR certified
Implementation & Tools Integration

Ready to implement revolutionary agent-API integration? Here are the tools, examples, and integration patterns to get started immediately.

🔧 Development Tools

Complete MCP API specification

JSON schema for credential feeds

👀 Live Examples

Preview API Feeds

See credential flows in action

Sign Credentials

Cryptographic credential signing

🚀 Quick Integration Pattern

📋 Agent Service Integration Logic🛡️ Template
Production integration pattern
// Agent discovers and validates API access\nasync function integrateWithService(serviceUrl, intent) {\n  // 1. Discover MCP feed\n  const mcpFeed = await fetch(`${serviceUrl}/.well-known/mcp.llmfeed.json`);\n  \n  // 2. Check if intent is supported\n  const capabilities = mcpFeed.capabilities.filter(c => c.intent === intent);\n  if (!capabilities.length) {\n    throw new Error(`Service doesn't support intent: ${intent}`);\n  }\n  \n  // 3. Get user credential\n\n// ... (content truncated for display - use copy button for full content)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
+11 more lines (use copy button for full content)
Join the Agent-API Revolution

The agentic web needs progressive autonomy and cryptographic trust. Help build the API integration infrastructure that will power autonomous agents for the next decade.

", "markdown": "# 🔐 API Integration Explained\n\nRevolutionary agent-native web integration with MCP protocol\n\n3\n\nAutonomy Levels\n\n100%\n\nSelf-Verifying\n\n∞\n\nAgent Delegation\n\nReal-time\n\nDiscovery\n\nRevolutionary Integration Guide\n\n🚀 Core Revolution\n\n* [• Progressive Autonomy](#progressive-autonomy)\n* [• Cryptographic Credentials](#cryptographic-credentials)\n* [• Agent Delegation](#agent-delegation)\n\n⚡ Integration Patterns\n\n* [• OpenAPI Bridge](#openapi-bridge)\n* [• Credential Bundles](#credential-bundles)\n* [• Implementation Guide](#implementation)\n\nThe Agent-API Revolution— Beyond Traditional Integration\n\nTraditional API integration assumes human oversight at every step. LLMFeed's **revolutionary MCP protocol integration** enables **progressive agent autonomy** — from assisted discovery to fully autonomous credential management and service orchestration.\n\n### Progressive Autonomy\n\nFrom manual config to full agent independence\n\n### Cryptographic Trust\n\nSelf-verifying credential management\n\n### Agent Delegation\n\nSecure agent-to-agent workflows\n\n#### Building on MCP Excellence\n\nLLMFeed extends Anthropic's **outstanding Model Context Protocol** from local tool calling to web-scale service discovery. We enhance MCP's robust JSON-RPC foundation with cryptographic trust, progressive autonomy, and multi-LLM compatibility — creating the complete agent-service integration stack.\n\nProgressive Autonomy Revolution\n\nThe future of agent-API integration isn't binary — it's a **progressive journey** from human-guided discovery to fully autonomous service orchestration.\n\n#### Level 1: Agent-Assisted Discovery (2025)\n\n**Scenario:** User needs document analysis\n\n**Agent:** \"I found DocumentAI via LLMFeed discovery. Great capabilities and trust scores. Would you like me to help set up access?\"\n\n**User:** \"Yes, show me what's involved\"\n\n**Agent:** \"I'll guide you through secure credential setup...\"\n\n**Human Role:** Approval and credential management\n\n#### Level 2: Semi-Autonomous Access (2026)\n\n**Scenario:** Trusted service with budget constraints\n\n**Agent:** \"I can use DocumentAI (certified, $0.03/page). Your budget allows 500 pages. Proceed?\"\n\n**User:** \"Yes, go ahead\"\n\n**Agent:** \\*Autonomously manages credentials and processes documents\\*\n\n**Human Role:** Budget approval and oversight\n\n#### Level 3: Full Autonomy (2027+)\n\n**Scenario:** Complex multi-service workflow\n\n**Agent:** \\*Autonomously discovers services, negotiates access, orchestrates 5-step workflow, provides results\\*\n\n**User:** Receives completed analysis with full audit trail\n\n**Human Role:** Goal setting and result review\n\nCryptographic Credential Revolution\n\nTraditional API keys are fragile, unverifiable strings. LLMFeed introduces **signed credential capsules** with cryptographic integrity, autonomous verification, and secure transfer capabilities.\n\n#### Traditional API Keys\n\n📋 Fragile API Key Usage🛡️ Template\n\nLegacy approach - vulnerable to attacksCopy\n\n```\n# Fragile and unverifiable\nexport API_KEY=\"sk_live_abc123...\"\ncurl -H \"Authorization: Bearer $API_KEY\" \\\n https://api.example.com/data\n```\n\n* ❌ No ownership verification\n* ❌ No integrity checking\n* ❌ No context or scope\n* ❌ Unsafe transfer\n* ❌ No audit trail\n\n#### LLMFeed Credential Capsules\n\n📋 Secure Credential Capsule🛡️ Template\n\nReplace {{placeholders}} with your valuesCopy\n\n```\n{\n \"feed_type\": \"credential\",\n \"credential\": {\n \"key_hint\": \"{{sk_live_...def456}}\",\n \"mcp_api\": \"https://{{api.example.com}}/.well-known/mcp-api.llmfeed.json?key=...\",\n \"allowed_intents\": [\"read_analytics\"],\n \"expires_at\": \"{{2025-12-10T14:30:00Z}}\"\n },\n \"trust\": {\n \"signed_blocks\": [\"credential\", \"trust\"],\n \"algorithm\": \"ed25519\"\n },\n\n// ... (content truncated for display - use copy button for full content)\n```\n\n1\n\n2\n\n3\n\n4\n\n5\n\n6\n\n7\n\n8\n\n9\n\n10\n\n11\n\n12\n\n13\n\n14\n\n+5 more lines (use copy button for full content)\n\n* ✅ Cryptographic verification\n* ✅ Tamper detection\n* ✅ Full context included\n* ✅ Secure capsule transfer\n* ✅ Complete audit trail\n\n#### Autonomous Credential Management\n\n📋 Autonomous Credential Manager🛡️ Template\n\nProduction credential validation logicCopy\n\n```\n// Agents can autonomously verify and manage credentials\nclass AutonomousCredentialManager {\n async validateAndStore(credentialFeed) {\n // 1. Verify cryptographic signature\n const signatureValid = await this.verifySignature(\n credentialFeed.signature.value,\n credentialFeed.trust.public_key_hint\n );\n \n if (!signatureValid) {\n throw new Error(\"Credential signature invalid - potential tampering\");\n }\n \n // 2. Check expiry autonomously\n const now = new Date();\n\n// ... (content truncated for display - use copy button for full content)\n```\n\n1\n\n2\n\n3\n\n4\n\n5\n\n6\n\n7\n\n8\n\n9\n\n10\n\n11\n\n12\n\n13\n\n14\n\n15\n\n16\n\n17\n\n+13 more lines (use copy button for full content)\n\nAgent-to-Agent Delegation Revolution\n\nThe future is **multi-agent workflows**. LLMFeed enables secure agent-to-agent credential delegation with scoped permissions, audit trails, and automatic revocation.\n\n#### Delegation Flow Example\n\n1\n\nMarketing agent receives credential feed\n\n2\n\nAgent evaluates delegation rules for analytics task\n\n3\n\nAgent requests temporary token via `delegation_endpoint`\n\n4\n\nAnalytics specialist receives scoped credential\n\n5\n\nAll actions logged with complete delegation context\n\n#### 🔒 Secure Delegation Structure\n\n📋 Agent Delegation Configuration🛡️ Template\n\nConfigure delegation rules for your agentsCopy\n\n```\n{\n \"credential\": {\n \"delegation_enabled\": true,\n \"delegation_rules\": [\n {\n \"target_agent\": \"{{analytics.specialist.ai}}\",\n \"allowed_intents\": [\"read_reports\"],\n \"max_duration\": \"1h\",\n \"audit_trail\": true\n }\n ],\n \"delegation_endpoint\": \"https://{{api.example.com}}/delegate\"\n }\n}\n```\n\n#### 📊 Audit & Security\n\n* • **Scoped Permissions:** Delegate only necessary intents\n* • **Time Limits:** Automatic expiration prevents abuse\n* • **Audit Trail:** Every action logged with delegation context\n* • **Revocation:** Primary agent can revoke at any time\n* • **Chain Tracking:** Full delegation chain visibility\n\nOpenAPI Integration Bridge\n\nLLMFeed doesn't replace existing standards — it **enhances them**. Our OpenAPI bridge combines intent understanding with technical specification for the complete agent integration experience.\n\n#### Best of Both Worlds Integration\n\n📋 OpenAPI + MCP Bridge🛡️ Template\n\nHybrid integration patternCopy\n\n```\n{\n \"capabilities\": [\n {\n \"type\": \"endpoint\",\n \"intent\": \"analyze document\",\n \"description\": \"AI-powered document analysis\", \n \"method\": \"POST\",\n \"path\": \"/api/analyze\",\n \"user_benefit\": \"Accurate OCR with 50+ language support\"\n },\n {\n \"type\": \"openapi\",\n \"url\": \"/.well-known/openapi.json\",\n \"description\": \"Complete technical specification\"\n }\n ]\n}\n```\n\n1\n\n2\n\n3\n\n4\n\n5\n\n6\n\n7\n\n8\n\n9\n\n10\n\n11\n\n12\n\n13\n\n14\n\n15\n\n16\n\n17\n\n→ **Agents understand intent** via LLMFeed, **validate parameters** via OpenAPI, **integrate via MCP** patterns.\n\n#### 🔧 Standards Compatibility\n\n* • **OpenAPI 3.1:** Technical specification\n* • **OAuth 2.0:** Authorization framework\n* • **JWT:** Secure credential transfer\n* • **Well-Known URIs:** Discovery patterns\n\n#### 🌉 Bridge Benefits\n\n* • Existing APIs work immediately\n* • No breaking changes required\n* • Progressive enhancement path\n* • Agent and human compatible\n\n#### ⚡ Enhanced Features\n\n* • Intent-based discovery\n* • Cryptographic trust\n* • Progressive autonomy\n* • Agent delegation support\n\nCredential Bundle Management\n\nModern workflows require **multiple services**. Credential bundles provide unified authentication, cross-service analytics, and coordinated billing for complex agent workflows.\n\n#### Marketing Suite Bundle Example\n\n📋 Multi-Service Credential Bundle🛡️ Template\n\nEnterprise template - adapt to your servicesCopy\n\n```\n{\n \"feed_type\": \"credential_bundle\",\n \"metadata\": {\n \"title\": \"{{Marketing Suite Access}}\",\n \"origin\": \"https://{{platform.example.com}}\"\n },\n \"credentials\": [\n {\n \"service\": \"email_api\",\n \"key_hint\": \"{{email_...x3f9}}\",\n \"mcp_api\": \"https://{{email.example.com}}/.well-known/mcp-api.llmfeed.json\"\n },\n {\n \"service\": \"analytics_api\", \n \"key_hint\": \"{{anl_...k7m2}}\",\n\n// ... (content truncated for display - use copy button for full content)\n```\n\n1\n\n2\n\n3\n\n4\n\n5\n\n6\n\n7\n\n8\n\n9\n\n10\n\n11\n\n12\n\n13\n\n14\n\n15\n\n16\n\n17\n\n+6 more lines (use copy button for full content)\n\n#### 💰 Enterprise Features\n\n* • **Usage-based billing:** $0.001/request\n* • **Included quota:** 50,000 requests\n* • **Overage handling:** Automatic rate adjustment\n* • **Cross-service optimization:** Bulk discounts\n\n#### 🏢 SSO Integration\n\n* • **Enterprise SSO:** Okta, Azure AD\n* • **Service accounts:** Automated workflows\n* • **Session management:** 8h duration\n* • **Compliance:** SOC2, GDPR certified\n\nImplementation & Tools Integration\n\nReady to implement revolutionary agent-API integration? Here are the tools, examples, and integration patterns to get started immediately.\n\n#### 🔧 Development Tools\n\nComplete MCP API specification\n\nJSON schema for credential feeds\n\n#### 👀 Live Examples\n\n[Preview API Feeds](/llmfeedhub/preview)\n\nSee credential flows in action\n\n[Sign Credentials](/tools/sign-and-verify)\n\nCryptographic credential signing\n\n#### 🚀 Quick Integration Pattern\n\n📋 Agent Service Integration Logic🛡️ Template\n\nProduction integration patternCopy\n\n```\n// Agent discovers and validates API access\nasync function integrateWithService(serviceUrl, intent) {\n // 1. Discover MCP feed\n const mcpFeed = await fetch(`${serviceUrl}/.well-known/mcp.llmfeed.json`);\n \n // 2. Check if intent is supported\n const capabilities = mcpFeed.capabilities.filter(c => c.intent === intent);\n if (!capabilities.length) {\n throw new Error(`Service doesn't support intent: ${intent}`);\n }\n \n // 3. Get user credential\n\n// ... (content truncated for display - use copy button for full content)\n```\n\n1\n\n2\n\n3\n\n4\n\n5\n\n6\n\n7\n\n8\n\n9\n\n10\n\n11\n\n12\n\n13\n\n14\n\n+11 more lines (use copy button for full content)\n\n#### 📚 Core Specs\n\n* [• Credential Feed Type](/spec/02_llmfeed_feedtype/llmfeed_feedtypes_credential)\n* [• API Extension Spec](/spec/03_llmfeed_extensions/llmfeed_extensions_api)\n* [• Agent Behavior Guidelines](/spec/04_agent-behavior/agent-behaviour)\n\n#### 🛠️ Tools\n\n* [• Sign & Verify](/tools/sign-and-verify)\n* [• Trust Verification](/verify)\n* [• Feed Hub](/llmfeedhub)\n\n#### 🌐 Standards\n\n* [• Anthropic MCP](https://modelcontextprotocol.io)\n* [• OpenAPI 3.1](https://spec.openapis.org/oas/v3.1.0)\n* [• OAuth 2.0](https://tools.ietf.org/html/rfc6749)\n\nJoin the Agent-API Revolution\n\nThe agentic web needs **progressive autonomy** and **cryptographic trust**. Help build the API integration infrastructure that will power autonomous agents for the next decade.\n\n[Join MCP Innovation](/join)[Start Signing Credentials](/tools/sign-and-verify)\n\nExport LLMFeed", "title": "🔐 API Integration Explained | Agent-Native Web Revolution | WellKnownMCP", "word_count": 1331, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of api-explained for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "API integration patterns with LLMFeed including authentication and agent-API protocols.", "tags": [ "api", "integration", "authentication", "patterns" ], "description_llm": "Technical guide for LLMFeed-compatible API integration", "intent": "integrate", "importance": "high", "audience": [ "developer", "api-architect" ] }, { "block_name": "app-mobile-explained", "type": "html+markdown", "timestamp": "2026-07-20T19:10:41.369089Z", "content": { "html": "

📱 Mobile App Explained

The agent-aware mobile revolution — from black boxes to intelligent interfaces

5
Implementation Phases
100%
Voice Control
9
Step Journey
∞
App Capabilities
Agent-Aware Mobile Revolution Guide
The Agent-Aware Mobile Revolution— From Black Boxes to Intelligent Interfaces

Mobile apps today are black boxes to AI agents — they can't be discovered, understood, or controlled conversationally. LLMFeed's revolutionary mobile app integration transforms applications into agent-aware, voice-controllable capabilities that work seamlessly with AI assistants.

Voice-First Control

Natural conversation controls all app functions

Progressive Discovery

Agent-guided journey from discovery to mastery

Cross-Platform Bridge

Seamless web-to-mobile capability transfer

The Current Mobile App Problem

Apps Are Black Boxes:

  • • Agents can't discover app capabilities
  • • No way to understand app functions
  • • Manual navigation required
  • • Voice assistants limited to app launching

Agent-Aware Solution:

  • • Apps declare capabilities for agent discovery
  • • Voice control of all app functions
  • • Agent-guided installation and setup
  • • Cross-platform workflow integration
The Invisible App Revolution

The future isn't about opening apps and tapping screens — it's about apps becoming invisible capability providers that agents orchestrate through natural conversation.

From Interface-First to Capability-First

Traditional Apps (Interface-First):

1. User opens app manually

2. Navigates through interface

3. Taps buttons and fills forms

4. Gets result after manual work

Agent-Aware Apps (Capability-First):

1. User speaks intent: \"Log my 5-mile run\"

2. Agent discovers app capability via LLMFeed

3. Agent calls app API with context

4. Result delivered conversationally

🎯 Revolutionary Example: Fitness App

User:

\"Hey Claude, I just finished a 5-mile run in 35 minutes\"

Agent (Invisible Actions):

  • • Discovers FitnessApp via /.well-known/mobile-app.llmfeed.json
  • • Calls logWorkout API with run data
  • • Retrieves personal best comparison
  • • Updates user's fitness goals progress

Agent Response:

\"Great job! I've logged your run. That's a new personal best pace! You're 2 runs away from your weekly goal.\"

→ Everything handled seamlessly via API → User gets value without touching phone

Apps become capability providers rather than user interfaces

Voice-Controlled App Interactions

Transform any mobile app into a voice-first experience where users control all functionality through natural conversation with AI agents.

Mobile App LLMFeed Declaration

📋 Voice-Aware Mobile App Configuration🛡️ Template
Replace {{placeholders}} with your app values
{\n  \"feed_type\": \"mobile-app\",\n  \"metadata\": {\n    \"title\": \"{{HealthSync Pro}}\",\n    \"app_name\": \"{{HealthSync}}\",\n    \"platform\": [\"ios\", \"android\"],\n    \"app_id\": {\n      \"ios\": \"{{com.healthsync.app}}\",\n      \"android\": \"{{com.healthsync.android}}\"\n    }\n  },\n  \"voice_capabilities\": [\n    {\n      \"intent\": \"track_sleep\",\n      \"voice_triggers\": [\"track my sleep\", \"log sleep\", \"sleep tracking\"],\n\n// ... (content truncated for display - use copy button for full content)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
+13 more lines (use copy button for full content)

🎤 Voice Interaction Examples

Health & Fitness:

  • • \"Log my 5-mile run\"
  • • \"How did I sleep last night?\"
  • • \"Start my morning workout\"
  • • \"What's my heart rate?\"

Productivity:

  • • \"Add meeting with John tomorrow\"
  • • \"Create shopping list for dinner\"
  • • \"Set reminder for doctor's appointment\"

⚡ Agent Behavior Guidelines

On App Discovery:

  • • Present available intents as voice options
  • • Explain app capabilities conversationally
  • • Offer guided voice command setup

During Interaction:

  • • Route voice prompts to matching app intents
  • • Provide contextual responses
  • • Handle errors gracefully with alternatives

🔄 Cross-App Voice Workflows

User: \"Plan my day\"

Agent: Queries calendar app, weather app, fitness app

Response: \"You have 3 meetings, 20% rain chance, and your workout goal needs 30 minutes. Should I schedule your run for 6 PM?\"

Progressive App Discovery Journey

Revolutionary agent-guided journey from app discovery to full voice control mastery — no app store confusion, no setup friction, no abandoned installations.

9-Step Agent-Guided Journey

1

Discovery:

User asks ChatGPT \"Help me track my fitness\"

2

Agent Discovery:

Agent finds healthsync.com/.well-known/mobile-app.llmfeed.json

3

Explanation:

\"I found HealthSync - it has AI coaching and voice tracking\"

4

Demo:

Agent shows web demo of workout planning

5

Configuration:

Agent asks preferences while user is engaged

6

Installation:

\"Should I open the App Store for you?\"

7

Handoff:

App opens with pre-configured settings

8

Integration:

\"Try saying 'log my workout' to test voice commands\"

9

Success:

User has fully configured, agent-ready health app

💎 Value Proposition Matrix

StakeholderBenefit
UsersSeamless onboarding, pre-configured apps, educated about features
App DevelopersHigher conversion rates, better user activation, agent-driven growth
AgentsCan recommend and set up mobile apps intelligently
EcosystemBridge between web agents and mobile capabilities
Cross-Platform Capability Bridge

Revolutionary progressive enhancement system that bridges web and mobile capabilities, providing graceful fallbacks and intelligent capability orchestration.

Capability Bridge Architecture

📋 Cross-Platform Bridge Configuration🛡️ Template
Progressive enhancement pattern
{\n  \"capability_bridge\": {\n    \"web_fallbacks\": [\n      {\n        \"mobile_capability\": \"logWorkout\", \n        \"web_equivalent\": \"/dashboard/log-workout\",\n        \"agent_explanation\": \"You can log workouts here on web, then sync to mobile later\"\n      },\n      {\n        \"mobile_capability\": \"viewStats\",\n        \"web_equivalent\": \"/dashboard/statistics\", \n        \"real_time_sync\": true\n\n// ... (content truncated for display - use copy button for full content)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
+9 more lines (use copy button for full content)

🌐 Web-Only Experience

  • • Basic Tracking: Manual data entry
  • • Web Dashboard: View statistics and history
  • • Agent Integration: Web-based voice commands
  • • Demo Mode: Full feature preview

📱 Web + Mobile

  • • Advanced Tracking: Automatic data collection
  • • Voice Control: Hands-free app interaction
  • • Real-time Sync: Seamless data continuity
  • • Context Awareness: Location and activity-based features

⌚ Mobile + Wearables

  • • Comprehensive Ecosystem: Full health monitoring
  • • Automatic Data: Passive health tracking
  • • AI Insights: Predictive health recommendations
  • • Ecosystem Integration: All devices work together

🔄 Agent Configuration Flow

📋 Agent Pre-Install Configuration🛡️ Template
Configure agent-guided app setup
{\n  \"agent_configuration\": {\n    \"pre_install_setup\": [\n      {\n        \"step\": \"preferences_gathering\",\n        \"agent_questions\": [\n          \"What are your main fitness goals?\",\n          \"Do you prefer morning or evening workouts?\", \n          \"Would you like me to set up voice commands?\"\n        ],\n        \"storage\": \"temporary_session_for_app_handoff\"\n      }\n    ],\n    \"post_install_handoff\": {\n      \"deep_link\": \"{{healthsync://onboarding/agent-configured?session={session_id}}}\",\n      \"data_transfer\": \"encrypted_preferences_bundle\",\n      \"agent_introduction\": \"I've pre-configured your settings. Try saying 'Hey Siri, log my workout' to test voice commands.\"\n    }\n  }\n}
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Complete Web-to-Mobile User Journey

Experience the complete transformation from traditional app discovery to agent-orchestrated mobile experiences that feel like magic.

Traditional App Discovery

1. User searches app store blindly

2. Downloads random fitness app

3. Struggles with complex onboarding

4. Abandons app after confusion

5. Never achieves fitness goals

Agent-Guided Discovery

1. Natural conversation: \"Help me get fit\"

2. Agent finds perfect app match

3. Guided installation with pre-config

4. Voice control setup and training

5. Achieves goals with agent support

📈 Installation Analytics & Optimization

📋 Agent-Driven Analytics Configuration🛡️ Template
Analytics and optimization setup
{\n  \"installation_analytics\": {\n    \"agent_attribution\": true,\n    \"conversion_tracking\": [\n      \"agent_interaction_to_install\",\n      \"web_demo_to_install\", \n      \"configuration_completion_rate\"\n    ],\n    \"optimization_metrics\": [\n      \"dialogue_effectiveness\",\n      \"user_satisfaction_post_install\",\n      \"feature_adoption_rate\"\n    ],\n    \"privacy_compliant\": true,\n    \"gdpr_ready\": true\n  }\n}

App developers get unprecedented insights into agent-driven installations and user success patterns.

From App Store Confusion to Agent-Orchestrated Mobile Magic

Implementation Roadmap & Tools

The mobile agent revolution unfolds in five phases, from basic API bridges to full voice-first, seamless app interactions.

Phase 1: Add API Bridge (2025)

Add API bridge to existing mobile-app feeds for basic agent interaction

Phase 2: Agent Authentication (2026)

Build agent authentication flows for secure app access and user consent

Phase 3: Conversational Responses (2027)

Create conversational response formats for natural agent-app communication

Phase 4: Cross-App Workflows (2028)

Enable cross-app agent workflows with coordinated multi-app interactions

Phase 5: Full Voice-First Experience (2029)

Complete invisible app revolution with seamless voice-first mobile experiences

🛠️ For App Developers

Implementation Steps:

  • • Add /.well-known/mobile-app.llmfeed.json to website
  • • Create web demos of key mobile features
  • • Build configuration API for pre-install setup
  • • Add agent handoff deep links to mobile app
  • • Track agent-driven installs for optimization

🤖 For Web Agents

Integration Guidelines:

  • • Check for mobile-app feeds when users ask about functionality
  • • Offer app installation help with context and value explanation
  • • Use lite capabilities to demonstrate value before install
  • • Handle configuration dialogue to reduce onboarding friction
  • • Provide post-install support via voice commands

🚀 Development Tools & Examples

Complete mobile app feed example

Preview Mobile Integration

See agent-app interactions

Join the Mobile Agent Revolution

The future is voice-first, agent-orchestrated mobile experiences. Build the apps that work seamlessly with AI assistants and give users magical conversational control over their mobile world.

Apps become invisible capability providers. Users get magical experiences.

The next generation of mobile apps will be discovered, installed, configured, and controlled entirely through conversation.

", "markdown": "# 📱 Mobile App Explained\n\nThe agent-aware mobile revolution — from black boxes to intelligent interfaces\n\n5\n\nImplementation Phases\n\n100%\n\nVoice Control\n\n9\n\nStep Journey\n\n∞\n\nApp Capabilities\n\nAgent-Aware Mobile Revolution Guide\n\n🚀 Core Revolution\n\n* [• The Invisible App Revolution](#invisible-revolution)\n* [• Voice-Controlled Interactions](#voice-control)\n* [• Progressive App Discovery](#progressive-discovery)\n\n⚡ Advanced Features\n\n* [• Cross-Platform Bridge](#capability-bridge)\n* [• Complete User Journey](#user-journey)\n* [• Implementation Roadmap](#implementation)\n\nThe Agent-Aware Mobile Revolution— From Black Boxes to Intelligent Interfaces\n\nMobile apps today are **black boxes to AI agents** — they can't be discovered, understood, or controlled conversationally. LLMFeed's revolutionary mobile app integration transforms applications into **agent-aware, voice-controllable capabilities** that work seamlessly with AI assistants.\n\n### Voice-First Control\n\nNatural conversation controls all app functions\n\n### Progressive Discovery\n\nAgent-guided journey from discovery to mastery\n\n### Cross-Platform Bridge\n\nSeamless web-to-mobile capability transfer\n\n#### The Current Mobile App Problem\n\nApps Are Black Boxes:\n\n* • Agents can't discover app capabilities\n* • No way to understand app functions\n* • Manual navigation required\n* • Voice assistants limited to app launching\n\nAgent-Aware Solution:\n\n* • Apps declare capabilities for agent discovery\n* • Voice control of all app functions\n* • Agent-guided installation and setup\n* • Cross-platform workflow integration\n\nThe Invisible App Revolution\n\nThe future isn't about opening apps and tapping screens — it's about **apps becoming invisible capability providers** that agents orchestrate through natural conversation.\n\n#### From Interface-First to Capability-First\n\nTraditional Apps (Interface-First):\n\n1. User opens app manually\n\n2. Navigates through interface\n\n3. Taps buttons and fills forms\n\n4. Gets result after manual work\n\nAgent-Aware Apps (Capability-First):\n\n1. User speaks intent: \"Log my 5-mile run\"\n\n2. Agent discovers app capability via LLMFeed\n\n3. Agent calls app API with context\n\n4. Result delivered conversationally\n\n#### 🎯 Revolutionary Example: Fitness App\n\nUser:\n\n\"Hey Claude, I just finished a 5-mile run in 35 minutes\"\n\nAgent (Invisible Actions):\n\n* • Discovers FitnessApp via `/.well-known/mobile-app.llmfeed.json`\n* • Calls `logWorkout` API with run data\n* • Retrieves personal best comparison\n* • Updates user's fitness goals progress\n\nAgent Response:\n\n\"Great job! I've logged your run. That's a new personal best pace! You're 2 runs away from your weekly goal.\"\n\n→ Everything handled seamlessly via API → User gets value without touching phone\n\nApps become **capability providers** rather than **user interfaces**\n\nVoice-Controlled App Interactions\n\nTransform any mobile app into a **voice-first experience** where users control all functionality through natural conversation with AI agents.\n\n#### Mobile App LLMFeed Declaration\n\n📋 Voice-Aware Mobile App Configuration🛡️ Template\n\nReplace {{placeholders}} with your app valuesCopy\n\n```\n{\n \"feed_type\": \"mobile-app\",\n \"metadata\": {\n \"title\": \"{{HealthSync Pro}}\",\n \"app_name\": \"{{HealthSync}}\",\n \"platform\": [\"ios\", \"android\"],\n \"app_id\": {\n \"ios\": \"{{com.healthsync.app}}\",\n \"android\": \"{{com.healthsync.android}}\"\n }\n },\n \"voice_capabilities\": [\n {\n \"intent\": \"track_sleep\",\n \"voice_triggers\": [\"track my sleep\", \"log sleep\", \"sleep tracking\"],\n\n// ... (content truncated for display - use copy button for full content)\n```\n\n1\n\n2\n\n3\n\n4\n\n5\n\n6\n\n7\n\n8\n\n9\n\n10\n\n11\n\n12\n\n13\n\n14\n\n15\n\n16\n\n17\n\n+13 more lines (use copy button for full content)\n\n#### 🎤 Voice Interaction Examples\n\nHealth & Fitness:\n\n* • \"Log my 5-mile run\"\n* • \"How did I sleep last night?\"\n* • \"Start my morning workout\"\n* • \"What's my heart rate?\"\n\nProductivity:\n\n* • \"Add meeting with John tomorrow\"\n* • \"Create shopping list for dinner\"\n* • \"Set reminder for doctor's appointment\"\n\n#### ⚡ Agent Behavior Guidelines\n\nOn App Discovery:\n\n* • Present available intents as voice options\n* • Explain app capabilities conversationally\n* • Offer guided voice command setup\n\nDuring Interaction:\n\n* • Route voice prompts to matching app intents\n* • Provide contextual responses\n* • Handle errors gracefully with alternatives\n\n#### 🔄 Cross-App Voice Workflows\n\n**User:** \"Plan my day\"\n\n**Agent:** *Queries calendar app, weather app, fitness app*\n\n**Response:** \"You have 3 meetings, 20% rain chance, and your workout goal needs 30 minutes. Should I schedule your run for 6 PM?\"\n\nProgressive App Discovery Journey\n\nRevolutionary **agent-guided journey** from app discovery to full voice control mastery — no app store confusion, no setup friction, no abandoned installations.\n\n#### 9-Step Agent-Guided Journey\n\n1\n\nDiscovery:\n\nUser asks ChatGPT \"Help me track my fitness\"\n\n2\n\nAgent Discovery:\n\nAgent finds `healthsync.com/.well-known/mobile-app.llmfeed.json`\n\n3\n\nExplanation:\n\n\"I found HealthSync - it has AI coaching and voice tracking\"\n\n4\n\nDemo:\n\nAgent shows web demo of workout planning\n\n5\n\nConfiguration:\n\nAgent asks preferences while user is engaged\n\n6\n\nInstallation:\n\n\"Should I open the App Store for you?\"\n\n7\n\nHandoff:\n\nApp opens with pre-configured settings\n\n8\n\nIntegration:\n\n\"Try saying 'log my workout' to test voice commands\"\n\n9\n\nSuccess:\n\nUser has fully configured, agent-ready health app\n\n#### 💎 Value Proposition Matrix\n\n| Stakeholder | Benefit |\n| --- | --- |\n| **Users** | Seamless onboarding, pre-configured apps, educated about features |\n| **App Developers** | Higher conversion rates, better user activation, agent-driven growth |\n| **Agents** | Can recommend and set up mobile apps intelligently |\n| **Ecosystem** | Bridge between web agents and mobile capabilities |\n\nCross-Platform Capability Bridge\n\nRevolutionary **progressive enhancement** system that bridges web and mobile capabilities, providing graceful fallbacks and intelligent capability orchestration.\n\n#### Capability Bridge Architecture\n\n📋 Cross-Platform Bridge Configuration🛡️ Template\n\nProgressive enhancement patternCopy\n\n```\n{\n \"capability_bridge\": {\n \"web_fallbacks\": [\n {\n \"mobile_capability\": \"logWorkout\", \n \"web_equivalent\": \"/dashboard/log-workout\",\n \"agent_explanation\": \"You can log workouts here on web, then sync to mobile later\"\n },\n {\n \"mobile_capability\": \"viewStats\",\n \"web_equivalent\": \"/dashboard/statistics\", \n \"real_time_sync\": true\n\n// ... (content truncated for display - use copy button for full content)\n```\n\n1\n\n2\n\n3\n\n4\n\n5\n\n6\n\n7\n\n8\n\n9\n\n10\n\n11\n\n12\n\n13\n\n14\n\n+9 more lines (use copy button for full content)\n\n#### 🌐 Web-Only Experience\n\n* • **Basic Tracking:** Manual data entry\n* • **Web Dashboard:** View statistics and history\n* • **Agent Integration:** Web-based voice commands\n* • **Demo Mode:** Full feature preview\n\n#### 📱 Web + Mobile\n\n* • **Advanced Tracking:** Automatic data collection\n* • **Voice Control:** Hands-free app interaction\n* • **Real-time Sync:** Seamless data continuity\n* • **Context Awareness:** Location and activity-based features\n\n#### ⌚ Mobile + Wearables\n\n* • **Comprehensive Ecosystem:** Full health monitoring\n* • **Automatic Data:** Passive health tracking\n* • **AI Insights:** Predictive health recommendations\n* • **Ecosystem Integration:** All devices work together\n\n#### 🔄 Agent Configuration Flow\n\n📋 Agent Pre-Install Configuration🛡️ Template\n\nConfigure agent-guided app setupCopy\n\n```\n{\n \"agent_configuration\": {\n \"pre_install_setup\": [\n {\n \"step\": \"preferences_gathering\",\n \"agent_questions\": [\n \"What are your main fitness goals?\",\n \"Do you prefer morning or evening workouts?\", \n \"Would you like me to set up voice commands?\"\n ],\n \"storage\": \"temporary_session_for_app_handoff\"\n }\n ],\n \"post_install_handoff\": {\n \"deep_link\": \"{{healthsync://onboarding/agent-configured?session={session_id}}}\",\n \"data_transfer\": \"encrypted_preferences_bundle\",\n \"agent_introduction\": \"I've pre-configured your settings. Try saying 'Hey Siri, log my workout' to test voice commands.\"\n }\n }\n}\n```\n\n1\n\n2\n\n3\n\n4\n\n5\n\n6\n\n7\n\n8\n\n9\n\n10\n\n11\n\n12\n\n13\n\n14\n\n15\n\n16\n\n17\n\n18\n\n19\n\n20\n\nComplete Web-to-Mobile User Journey\n\nExperience the **complete transformation** from traditional app discovery to agent-orchestrated mobile experiences that feel like magic.\n\n#### Traditional App Discovery\n\n1. User searches app store blindly\n\n2. Downloads random fitness app\n\n3. Struggles with complex onboarding\n\n4. Abandons app after confusion\n\n5. Never achieves fitness goals\n\n#### Agent-Guided Discovery\n\n1. Natural conversation: \"Help me get fit\"\n\n2. Agent finds perfect app match\n\n3. Guided installation with pre-config\n\n4. Voice control setup and training\n\n5. Achieves goals with agent support\n\n#### 📈 Installation Analytics & Optimization\n\n📋 Agent-Driven Analytics Configuration🛡️ Template\n\nAnalytics and optimization setupCopy\n\n```\n{\n \"installation_analytics\": {\n \"agent_attribution\": true,\n \"conversion_tracking\": [\n \"agent_interaction_to_install\",\n \"web_demo_to_install\", \n \"configuration_completion_rate\"\n ],\n \"optimization_metrics\": [\n \"dialogue_effectiveness\",\n \"user_satisfaction_post_install\",\n \"feature_adoption_rate\"\n ],\n \"privacy_compliant\": true,\n \"gdpr_ready\": true\n }\n}\n```\n\nApp developers get unprecedented insights into agent-driven installations and user success patterns.\n\nFrom App Store Confusion to Agent-Orchestrated Mobile Magic\n\nImplementation Roadmap & Tools\n\nThe mobile agent revolution unfolds in **five phases**, from basic API bridges to full voice-first, seamless app interactions.\n\n#### Phase 1: Add API Bridge (2025)\n\nAdd API bridge to existing mobile-app feeds for basic agent interaction\n\n#### Phase 2: Agent Authentication (2026)\n\nBuild agent authentication flows for secure app access and user consent\n\n#### Phase 3: Conversational Responses (2027)\n\nCreate conversational response formats for natural agent-app communication\n\n#### Phase 4: Cross-App Workflows (2028)\n\nEnable cross-app agent workflows with coordinated multi-app interactions\n\n#### Phase 5: Full Voice-First Experience (2029)\n\nComplete invisible app revolution with seamless voice-first mobile experiences\n\n#### 🛠️ For App Developers\n\nImplementation Steps:\n\n* • Add `/.well-known/mobile-app.llmfeed.json` to website\n* • Create web demos of key mobile features\n* • Build configuration API for pre-install setup\n* • Add agent handoff deep links to mobile app\n* • Track agent-driven installs for optimization\n\n#### 🤖 For Web Agents\n\nIntegration Guidelines:\n\n* • Check for mobile-app feeds when users ask about functionality\n* • Offer app installation help with context and value explanation\n* • Use lite capabilities to demonstrate value before install\n* • Handle configuration dialogue to reduce onboarding friction\n* • Provide post-install support via voice commands\n\n#### 🚀 Development Tools & Examples\n\nComplete mobile app feed example\n\n[Preview Mobile Integration](/llmfeedhub/preview)\n\nSee agent-app interactions\n\n#### 📚 Specifications\n\n* [• Mobile App Feed Type](/spec/02_llmfeed_feedtype/llmfeed_feedtype_mobile-app)\n* [• Agent Behavior Guidelines](/spec/04_agent-behavior/agent-behavior)\n* [• Privacy Extensions](/spec/03_llmfeed_extensions/llmfeed_extensions_privacy)\n\n#### 🛠️ Tools\n\n* [• Sign & Verify](/tools/sign-and-verify)\n* [• Trust Verification](/verify)\n* [• Feed Hub](/llmfeedhub)\n\n#### 🌐 Related\n\n* [• API Integration](/tools/api-explained)\n* [• Capabilities](/tools/capabilities-explained)\n* [• User Spaces](/tools/user-spaces-explained)\n\nJoin the Mobile Agent Revolution\n\nThe future is **voice-first, agent-orchestrated mobile experiences**. Build the apps that work seamlessly with AI assistants and give users magical conversational control over their mobile world.\n\nApps become invisible capability providers. Users get magical experiences.\n\n**The next generation of mobile apps will be discovered, installed, configured, and controlled entirely through conversation.**\n\n[Join Mobile Revolution](/join)[Start Building](/tools/sign-and-verify)\n\nExport LLMFeed", "title": "📱 Mobile App Explained | Agent-Aware App Revolution | WellKnownMCP", "word_count": 1591, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of app-mobile-explained for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Mobile app functionality exposure to agents via intent handling and deep linking.", "tags": [ "mobile", "app", "intents", "deep-linking" ], "description_llm": "Mobile application agent integration guide", "intent": "integrate", "importance": "medium", "audience": [ "mobile-developer" ] }, { "block_name": "badges", "type": "html+markdown", "timestamp": "2026-07-20T19:10:41.396887Z", "content": { "html": "

🔖 MCP + LLMFeed Badges

Visual trust and capability indicators for the agentic web

0

Total Badges

0

Ready

0

Verified

0

Signed

0

Certified

What are MCP badges?

Visual indicators that show your website's trust level, export capabilities, and agent compatibility at a glance.

Types: Protocol scope (MCP, LLMFeed, Export)
Levels: Trust status (Ready → Certified)
Visual: Unique emoji and color combinations
Embed: Copy-paste ready HTML snippets
Embed a Badge

Badge Gallery

Download All Badges

Get all badges + metadata for offline use or integration into your build pipeline.

NEW

", "markdown": "# 🔖 MCP + LLMFeed Badges\n\nVisual trust and capability indicators for the agentic web\n\n0\n\nTotal Badges\n\n0\n\nReady\n\n0\n\nVerified\n\n0\n\nSigned\n\n0\n\nCertified\n\n### What are MCP badges?\n\nVisual indicators that show your website's trust level, export capabilities, and agent compatibility at a glance.\n\n**Types:** Protocol scope (MCP, LLMFeed, Export)\n\n**Levels:** Trust status (Ready → Certified)\n\n**Visual:** Unique emoji and color combinations\n\n**Embed:** Copy-paste ready HTML snippets\n\nEmbed a Badge\n\nSelect Badge:-- Choose a badge --\n\nCopy Embed Code\n\n## Badge Gallery\n\nDownload All Badges\n\nGet all badges + metadata for offline use or integration into your build pipeline.\n\n[Download SVG ZIP](/assets/badges/combined_badges.zip)[Download .llmfeed.json](/assets/badges/badges.export.llmfeed.json)\n\n[X](https://twitter.com/intent/tweet?text=Discover%20this%20MCP-enabled%20resource&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fbadges \"Share on X\")[LinkedIn](https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fbadges&title=Discover%20this%20MCP-enabled%20resource \"Share on LinkedIn\")[WhatsApp](https://wa.me/?text=Discover%20this%20MCP-enabled%20resource%20https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fbadges \"Share on WhatsApp\")Copy link\n\nExport LLMFeedNEW", "title": "MCP + LLMFeed Badges — Trust & Capability Indicators", "word_count": 119, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of badges for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Visual trust and certification indicators with verification levels and display guidelines.", "tags": [ "badges", "trust", "certification", "visual" ], "description_llm": "Trust badges and certification display reference", "intent": "display", "importance": "medium", "audience": [ "webmaster", "designer" ] }, { "block_name": "feed-flagging", "type": "html+markdown", "timestamp": "2026-07-20T19:10:41.466484Z", "content": { "html": "

🚩 Feed Flagging Explained

Revolutionary decentralized trust management for the agentic web

3
Flagging Forms
4
Trust Levels
Real-time
AI Detection
∞
Democratic Scale
Complete Guide Navigation
The Flagging Revolution— Trust at Internet Scale

Traditional content moderation doesn't scale to the agentic web. LLMFeed's revolutionary flagging system combines AI-powered threat detection, cryptoeconomic incentives, and democratic governance to create the world's first decentralized trust management system for autonomous agents.

AI Detection

Real-time threat detection with machine learning

Democratic Governance

Community-driven trust decisions

Economic Incentives

Cryptoeconomic reputation mining

Three Forms of Flagging

The LLMFeed flagging system operates through three distinct but coordinated mechanisms, each optimized for different scenarios and use cases.

1

Embedded Flags

flags[] array directly in the .llmfeed.json file

\"flags\": [{\n  \"type\": \"risk\",\n  \"submitted_by\": \"agent://auditbot\",\n  \"reason\": \"Capability mismatch\",\n  \"status\": \"pending\",\n  \"date\": \"2025-06-19T12:00:00Z\"\n}]

✅ Immediate visibility | ⚠️ Requires author cooperation

2

Separate Flag Feeds

Independent flag.llmfeed.json published by auditors

{\n  \"feed_type\": \"flag\",\n  \"target\": \"https://evil.com/.well-known/mcp.llmfeed.json\",\n  \"reason\": \"Spoofed capabilities\",\n  \"submitted_by\": \"https://trustbot.ai\"\n}

✅ Independent review | ✅ No permission needed

3

LLMCA Submission

Central reporting to nonprofit governance entity

POST https://llmca.org/api/flags\n{\n  \"feed_url\": \"https://example.org/.well-known/mcp.llmfeed.json\",\n  \"violation_type\": \"trust_abuse\",\n  \"evidence\": {...}\n}

✅ Official authority | ✅ Appeals process

Real-Time Trust Scoring Revolution

The flagging system integrates with LLMFeed's 4-level trust scoring to provide real-time trust adjustment with millisecond response times.

Trust LevelFlagging ImpactResponse TimeAppeal Priority
UNTRUSTEDImmediate restrictionReal-time🚨 Emergency review
BASICCautionary warnings< 1 hour⚡ Standard process
VERIFIEDInvestigation required< 24 hours🔍 Thorough review
PREMIUMMultiple flags needed< 1 week⚖️ Due process guaranteed

Dynamic Response Example

{\n  \"dynamic_response\": {\n    \"threat_level_assessment\": {\n      \"imminent_harm\": \"immediate_suspension\",\n      \"reputation_gaming\": \"gradual_trust_reduction\", \n      \"technical_issues\": \"warning_with_investigation\"\n    },\n    \"escalation_triggers\": {\n      \"multiple_flags_24h\": \"auto_escalate_to_human_review\",\n      \"verified_agent_flag\": \"priority_investigation\", \n      \"community_consensus\": \"democratic_resolution\"\n    }\n  }\n}
AI-Powered Threat Detection Systems

Revolutionary AI detection capabilities include manipulation detection, content safety screening, and quality assurance — all running in real-time with human-AI collaboration frameworks.

🎯 Manipulation Detection

  • • Coordinated inauthentic behavior
  • • Reputation gaming patterns
  • • Sockpuppet network analysis
  • • Economic manipulation detection

🛡️ Content Safety

  • • Harmful guidance screening
  • • Privacy violation detection
  • • Security risk assessment
  • • Regulatory compliance checking

✅ Quality Assurance

  • • Technical accuracy validation
  • • Logical consistency verification
  • • Performance claims testing
  • • User experience scoring

Human-AI Collaboration Framework

AI Contributions:

  • • Pattern recognition at scale
  • • 24/7 continuous monitoring
  • • Statistical anomaly detection
  • • Multi-language analysis

Human Oversight:

  • • Contextual judgment calls
  • • Cultural sensitivity review
  • • Edge case resolution
  • • Appeals and corrections
Democratic Governance Revolution

Unlike centralized moderation, LLMFeed implements democratic governance with elected community councils, transparent appeals processes, and anti-corruption mechanisms.

🏛️ Flagging Council

  • • Multi-stakeholder elected representation
  • • Qualified majority with minority protection
  • • All decisions publicly auditable
  • • Regular elections and recall procedures

⚖️ Appeals Process

  • • Automated AI screening
  • • Diverse panel representation
  • • Community vote with veto protection
  • • Legal expert panel for edge cases

Anti-Corruption Mechanisms

Structural:

  • • Term limits for governance roles
  • • Separation of powers
  • • Checks and balances

Transparency:

  • • Cryptographic voting verification
  • • Public audit logs
  • • Real-time decision tracking

Protection:

  • • Whistleblower anonymity
  • • Minority veto power
  • • External audit requirements
Cryptoeconomic Incentives & Reputation Mining

Revolutionary reputation mining system where participants earn verifiable reputation through quality contributions, with sophisticated anti-gaming mechanisms.

🏆 Earning Reputation

  • • Accurate Flagging: Bonus for identifying real problems
  • • Quality Review: Rewards for thorough assessment
  • • Community Service: Governance participation recognition
  • • Whistleblowing: Protected rewards for serious violations

⚡ Anti-Gaming Protection

  • • Stake-weighted voting with anti-plutocracy
  • • Quadratic funding for community initiatives
  • • Sybil resistance through identity verification
  • • Long-term reputation staking requirements

💎 Economic Incentive Structure

{\n  \"incentive_structure\": {\n    \"flagging_rewards\": {\n      \"accurate_flags\": \"reputation_tokens_plus_monetary_bonus\",\n      \"false_positives\": \"reputation_penalty_with_learning_credit\",\n      \"malicious_flagging\": \"severe_reputation_loss_and_temporary_ban\"\n    },\n    \"governance_participation\": {\n      \"council_service\": \"governance_tokens_with_voting_power\",\n      \"community_voting\": \"participation_rewards_scaled_by_stake\",\n      \"appeal_review\": \"expert_witness_compensation\"\n    }\n  }\n}
Industry-Specific Adaptations

The flagging system adapts to industry-specific requirements with specialized compliance checking, ethics review, and regulatory monitoring.

🏥 Healthcare

  • • HIPAA compliance flagging
  • • Medical ethics review
  • • Clinical trial integrity
  • • Patient safety monitoring

🏦 Financial Services

  • • SOX/PCI DSS/GDPR flagging
  • • Market manipulation detection
  • • Fiduciary duty monitoring
  • • Systemic risk assessment

🏛️ Government

  • • Constitutional compliance
  • • Democratic oversight
  • • Transparency requirements
  • • Civil rights protection
Implementation & Tools Integration

Ready to implement the flagging system? Here are the tools, examples, and integration guides to get started immediately.

🔧 Development Tools

Agent behavior override prompts

Complete schema with flagging structure

👀 Live Examples

Preview Flagged Feed

See flagging system in action

Verify Trust Status

Check feed trust and flags

🚀 Quick Implementation Example

// Agent behavior when encountering flagged feeds\nif (feed.flags && feed.flags.length > 0) {\n  const criticalFlags = feed.flags.filter(f => f.type === 'critical');\n  \n  if (criticalFlags.length > 0) {\n    // Immediate rejection for critical flags\n    return { action: 'reject', reason: 'Critical trust violations detected' };\n  }\n  \n  // Warning for other flags\n  return { \n    action: 'warn_user', \n    message: `This feed has ${feed.flags.length} community flag(s). Proceed with caution.`\n  };\n}
Join the Trust Revolution

The agentic web needs decentralized trust management. Help build the infrastructure that will secure autonomous agents for the next decade.

", "markdown": "# 🚩 Feed Flagging Explained\n\nRevolutionary decentralized trust management for the agentic web\n\n3\n\nFlagging Forms\n\n4\n\nTrust Levels\n\nReal-time\n\nAI Detection\n\n∞\n\nDemocratic Scale\n\nComplete Guide Navigation\n\n🔍 Core Systems\n\n* [• Three Flagging Forms](#three-forms)\n* [• Real-Time Trust Scoring](#trust-scoring)\n* [• AI Detection Systems](#ai-detection)\n\n⚡ Advanced Features\n\n* [• Democratic Governance](#governance)\n* [• Cryptoeconomic Incentives](#economics)\n* [• Implementation Guide](#implementation)\n\nThe Flagging Revolution— Trust at Internet Scale\n\nTraditional content moderation doesn't scale to the agentic web. LLMFeed's **revolutionary flagging system** combines AI-powered threat detection, cryptoeconomic incentives, and democratic governance to create the world's first **decentralized trust management system** for autonomous agents.\n\n### AI Detection\n\nReal-time threat detection with machine learning\n\n### Democratic Governance\n\nCommunity-driven trust decisions\n\n### Economic Incentives\n\nCryptoeconomic reputation mining\n\nThree Forms of Flagging\n\nThe LLMFeed flagging system operates through three distinct but coordinated mechanisms, each optimized for different scenarios and use cases.\n\n1\n\n### Embedded Flags\n\n`flags[]` array directly in the .llmfeed.json file\n\n```\n\"flags\": [{\n \"type\": \"risk\",\n \"submitted_by\": \"agent://auditbot\",\n \"reason\": \"Capability mismatch\",\n \"status\": \"pending\",\n \"date\": \"2025-06-19T12:00:00Z\"\n}]\n```\n\n✅ Immediate visibility | ⚠️ Requires author cooperation\n\n2\n\n### Separate Flag Feeds\n\nIndependent `flag.llmfeed.json` published by auditors\n\n```\n{\n \"feed_type\": \"flag\",\n \"target\": \"https://evil.com/.well-known/mcp.llmfeed.json\",\n \"reason\": \"Spoofed capabilities\",\n \"submitted_by\": \"https://trustbot.ai\"\n}\n```\n\n✅ Independent review | ✅ No permission needed\n\n3\n\n### LLMCA Submission\n\nCentral reporting to nonprofit governance entity\n\n```\nPOST https://llmca.org/api/flags\n{\n \"feed_url\": \"https://example.org/.well-known/mcp.llmfeed.json\",\n \"violation_type\": \"trust_abuse\",\n \"evidence\": {...}\n}\n```\n\n✅ Official authority | ✅ Appeals process\n\nReal-Time Trust Scoring Revolution\n\nThe flagging system integrates with LLMFeed's **4-level trust scoring** to provide real-time trust adjustment with millisecond response times.\n\n| Trust Level | Flagging Impact | Response Time | Appeal Priority |\n| --- | --- | --- | --- |\n| **UNTRUSTED** | Immediate restriction | Real-time | 🚨 Emergency review |\n| **BASIC** | Cautionary warnings | < 1 hour | ⚡ Standard process |\n| **VERIFIED** | Investigation required | < 24 hours | 🔍 Thorough review |\n| **PREMIUM** | Multiple flags needed | < 1 week | ⚖️ Due process guaranteed |\n\n#### Dynamic Response Example\n\n```\n{\n \"dynamic_response\": {\n \"threat_level_assessment\": {\n \"imminent_harm\": \"immediate_suspension\",\n \"reputation_gaming\": \"gradual_trust_reduction\", \n \"technical_issues\": \"warning_with_investigation\"\n },\n \"escalation_triggers\": {\n \"multiple_flags_24h\": \"auto_escalate_to_human_review\",\n \"verified_agent_flag\": \"priority_investigation\", \n \"community_consensus\": \"democratic_resolution\"\n }\n }\n}\n```\n\nAI-Powered Threat Detection Systems\n\nRevolutionary AI detection capabilities include manipulation detection, content safety screening, and quality assurance — all running in real-time with **human-AI collaboration frameworks**.\n\n#### 🎯 Manipulation Detection\n\n* • Coordinated inauthentic behavior\n* • Reputation gaming patterns\n* • Sockpuppet network analysis\n* • Economic manipulation detection\n\n#### 🛡️ Content Safety\n\n* • Harmful guidance screening\n* • Privacy violation detection\n* • Security risk assessment\n* • Regulatory compliance checking\n\n#### ✅ Quality Assurance\n\n* • Technical accuracy validation\n* • Logical consistency verification\n* • Performance claims testing\n* • User experience scoring\n\n#### Human-AI Collaboration Framework\n\nAI Contributions:\n\n* • Pattern recognition at scale\n* • 24/7 continuous monitoring\n* • Statistical anomaly detection\n* • Multi-language analysis\n\nHuman Oversight:\n\n* • Contextual judgment calls\n* • Cultural sensitivity review\n* • Edge case resolution\n* • Appeals and corrections\n\nDemocratic Governance Revolution\n\nUnlike centralized moderation, LLMFeed implements **democratic governance** with elected community councils, transparent appeals processes, and anti-corruption mechanisms.\n\n#### 🏛️ Flagging Council\n\n* • Multi-stakeholder elected representation\n* • Qualified majority with minority protection\n* • All decisions publicly auditable\n* • Regular elections and recall procedures\n\n#### ⚖️ Appeals Process\n\n* • Automated AI screening\n* • Diverse panel representation\n* • Community vote with veto protection\n* • Legal expert panel for edge cases\n\n#### Anti-Corruption Mechanisms\n\nStructural:\n\n* • Term limits for governance roles\n* • Separation of powers\n* • Checks and balances\n\nTransparency:\n\n* • Cryptographic voting verification\n* • Public audit logs\n* • Real-time decision tracking\n\nProtection:\n\n* • Whistleblower anonymity\n* • Minority veto power\n* • External audit requirements\n\nCryptoeconomic Incentives & Reputation Mining\n\nRevolutionary **reputation mining** system where participants earn verifiable reputation through quality contributions, with sophisticated anti-gaming mechanisms.\n\n#### 🏆 Earning Reputation\n\n* • **Accurate Flagging:** Bonus for identifying real problems\n* • **Quality Review:** Rewards for thorough assessment\n* • **Community Service:** Governance participation recognition\n* • **Whistleblowing:** Protected rewards for serious violations\n\n#### ⚡ Anti-Gaming Protection\n\n* • Stake-weighted voting with anti-plutocracy\n* • Quadratic funding for community initiatives\n* • Sybil resistance through identity verification\n* • Long-term reputation staking requirements\n\n#### 💎 Economic Incentive Structure\n\n```\n{\n \"incentive_structure\": {\n \"flagging_rewards\": {\n \"accurate_flags\": \"reputation_tokens_plus_monetary_bonus\",\n \"false_positives\": \"reputation_penalty_with_learning_credit\",\n \"malicious_flagging\": \"severe_reputation_loss_and_temporary_ban\"\n },\n \"governance_participation\": {\n \"council_service\": \"governance_tokens_with_voting_power\",\n \"community_voting\": \"participation_rewards_scaled_by_stake\",\n \"appeal_review\": \"expert_witness_compensation\"\n }\n }\n}\n```\n\nIndustry-Specific Adaptations\n\nThe flagging system adapts to industry-specific requirements with specialized compliance checking, ethics review, and regulatory monitoring.\n\n#### 🏥 Healthcare\n\n* • HIPAA compliance flagging\n* • Medical ethics review\n* • Clinical trial integrity\n* • Patient safety monitoring\n\n#### 🏦 Financial Services\n\n* • SOX/PCI DSS/GDPR flagging\n* • Market manipulation detection\n* • Fiduciary duty monitoring\n* • Systemic risk assessment\n\n#### 🏛️ Government\n\n* • Constitutional compliance\n* • Democratic oversight\n* • Transparency requirements\n* • Civil rights protection\n\nImplementation & Tools Integration\n\nReady to implement the flagging system? Here are the tools, examples, and integration guides to get started immediately.\n\n#### 🔧 Development Tools\n\nAgent behavior override prompts\n\nComplete schema with flagging structure\n\n#### 👀 Live Examples\n\n[Preview Flagged Feed](/llmfeedhub/preview?demo=kungfu)\n\nSee flagging system in action\n\n[Verify Trust Status](/verify)\n\nCheck feed trust and flags\n\n#### 🚀 Quick Implementation Example\n\n```\n// Agent behavior when encountering flagged feeds\nif (feed.flags && feed.flags.length > 0) {\n const criticalFlags = feed.flags.filter(f => f.type === 'critical');\n \n if (criticalFlags.length > 0) {\n // Immediate rejection for critical flags\n return { action: 'reject', reason: 'Critical trust violations detected' };\n }\n \n // Warning for other flags\n return { \n action: 'warn_user', \n message: `This feed has ${feed.flags.length} community flag(s). Proceed with caution.`\n };\n}\n```\n\nJoin the Trust Revolution\n\nThe agentic web needs **decentralized trust management**. Help build the infrastructure that will secure autonomous agents for the next decade.\n\n[Join LLMCA Governance](/join)[Start Signing Feeds](/tools/sign-and-verify)\n\nExport LLMFeed", "title": "🚩 Feed Flagging Explained | Decentralized Trust Management Revolution | WellKnownMCP", "word_count": 1006, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of feed-flagging for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Community moderation system for reporting malicious feeds and trust reputation management.", "tags": [ "moderation", "flagging", "community", "trust" ], "description_llm": "Community governance and feed quality control", "intent": "moderate", "importance": "medium", "audience": [ "community", "moderator" ] }, { "block_name": "pricing", "type": "html+markdown", "timestamp": "2026-07-20T19:10:41.541825Z", "content": { "html": "

💰 Agent Pricing Explained

Revolutionary autonomous economy platform for intelligent agents

NEW

4
Development Phases
∞
Agent Tiers
Real-time
ML Optimization
2028+
Full Autonomy
Agent Economy Revolution Guide
The Agent Economy Revolution— First Autonomous Marketplace

Traditional pricing assumes human oversight and manual negotiations. LLMFeed introduces the world's first autonomous agent economy — where intelligent systems discover, negotiate, and transact value independently while maintaining fairness, transparency, and optimal outcomes for all participants.

Intelligence-Based Tiers

GPT-3.5 class to GPT-4+ specialists

Dynamic ML Pricing

Real-time optimization algorithms

Multi-Agent Coordination

Automatic workflow billing

Revolutionary Economic Paradigm

This isn't just pricing — it's the foundation of the first autonomous agent economy. Agents discover capabilities, negotiate fair prices, coordinate complex workflows, and settle payments automatically. The future where AI systems transact value independently while optimizing for human outcomes.

Agent Economy Platform Architecture

The LLMFeed agent economy platform enables autonomous economic interactions between intelligent agents with trust scoring, performance optimization, cross-agent billing, and real-time negotiation capabilities.

Platform Core Features

{\n  \"agent_economy\": {\n    \"enabled\": true,\n    \"trust_scoring\": true,\n    \"performance_optimization\": true,\n    \"cross_agent_billing\": true,\n    \"real_time_negotiation\": true\n  }\n}

🤖 Agent Intelligence Tiers

Basic Health Agent

Intelligence: GPT-3.5-class

Rate: $0.05/consultation

Medical AI Specialist

Intelligence: GPT-4-medical-tuned

Rate: $0.75/analysis + performance bonus

Certified Expert

Intelligence: Domain-specialized

Rate: Outcome-based pricing

📊 Economic Metrics

  • • Success Rate: % of successful outcomes
  • • User Satisfaction: Average rating last 100 transactions
  • • Efficiency Score: Cost per successful outcome
  • • Reliability: Uptime and response consistency
  • • Market Intelligence: AI-powered demand forecasting
  • • Competitive Analysis: Market position tracking
Dynamic ML Pricing Intelligence

Revolutionary machine learning pricing optimization using reinforcement learning algorithms that balance revenue, outcomes, and agent adoption in real-time.

ML Optimization Engine

{\n  \"dynamic_pricing\": {\n    \"ml_optimization\": {\n      \"enabled\": true,\n      \"algorithm\": \"reinforcement_learning_pricing\",\n      \"optimization_goals\": [\"revenue\", \"patient_outcomes\", \"agent_adoption\"],\n      \"update_frequency\": \"every_hour\",\n      \"a_b_testing\": true\n    }\n  }\n}

⚡ Real-Time Factors

Agent Demand Surge

Impact: High (±50% adjustment)

Example: Flu season increases triage demand

Emergency Priority

Impact: Critical (3x multiplier)

Trigger: Life-threatening symptoms detected

Performance Bonus

95% accuracy: 1.0x | 98%: 1.2x | 99%: 1.5x

Quality-based pricing adjustments

🎯 Pricing Models

Fixed Rate Model

\"pricing\": {\n  \"base_rate\": 0.10,\n  \"unit\": \"per_request\",\n  \"volume_discounts\": {\n    \"100_requests\": 0.08,\n    \"1000_requests\": 0.06\n  }\n}

Performance-Based Model

\"pricing\": {\n  \"model\": \"outcome_based\",\n  \"success_fee\": 2.50,\n  \"success_criteria\": [\n    \"user_goal_achievement\",\n    \"satisfaction_score_8+\"\n  ],\n  \"base_fee\": 0.25\n}
Multi-Agent Workflow Coordination

Complex tasks require multiple specialist agents working together. LLMFeed coordinates billing, performance guarantees, and automatic settlements across agent networks.

Healthcare Workflow Example

{\n  \"comprehensive_health_analysis\": {\n    \"workflow_id\": \"full_health_assessment\",\n    \"agents\": [\n      {\"role\": \"intake_specialist\", \"cost_share\": 15, \"max_cost\": 0.30},\n      {\"role\": \"diagnostic_ai\", \"cost_share\": 50, \"max_cost\": 1.00},\n      {\"role\": \"specialist_consultation\", \"cost_share\": 25, \"max_cost\": 0.50},\n      {\"role\": \"care_coordinator\", \"cost_share\": 10, \"max_cost\": 0.20}\n    ],\n    \"total_cost_estimate\": \"1.50-2.50\",\n    \"user_payment\": \"single_charge_to_user\",\n    \"agent_settlement\": \"automatic_smart_contract\"\n  }\n}

🏥 Healthcare Network

  • • Agents: Intake, Diagnostic, Treatment
  • • Model: Outcome-based pricing
  • • Metric: Patient health improvement
  • • Range: $2.00-8.00
  • • Integration: Insurance compatible

⚖️ Legal Research

  • • Agents: Research, Analysis, Writing
  • • Model: Performance-tiered
  • • Metrics: Legal accuracy, argument strength
  • • Simple: $5.00
  • • Complex: $50.00

💼 Financial Advisory

  • • Agents: Risk, Prediction, Optimization
  • • Model: Performance fee
  • • Structure: 20% of excess returns
  • • Compliance: SEC registered
  • • Standard: Fiduciary duty

💰 Revenue Sharing Models

Collaborative Model:

  • • Primary Agent: 50%
  • • Supporting Agents: 35%
  • • Platform Fee: 15%
  • • Performance Bonus Pool: 20%

Auction Model:

  • • Complex Case Bidding: Enabled
  • • Quality Weighting: 70%
  • • Price Weighting: 30%
  • • Bid Timeout: 60 seconds
Autonomous Agent Negotiations

The future is autonomous negotiations where agents negotiate prices, terms, and conditions automatically while maintaining fairness and transparency.

Negotiation Parameters

{\n  \"autonomous_negotiation\": {\n    \"enabled\": true,\n    \"max_price_variance\": \"±20%\",\n    \"negotiation_timeout\": \"30_seconds\",\n    \"ai_arbitration\": \"for_disputes\"\n  }\n}

🤝 Negotiation Process

1

Agent analyzes market rates and performance history

2

Proposes initial price within variance range

3

Counter-negotiations within 30-second window

4

AI arbitration resolves disputes automatically

⚖️ Fairness Mechanisms

  • • Price Variance Limits: ±20% of base rate
  • • Performance History: Quality-based adjustments
  • • Market Analysis: Competitive rate comparison
  • • User Benefit Priority: Optimize for outcomes
  • • Dispute Resolution: AI arbitration system
  • • Transparency: All negotiations logged

📊 Agent Behavior Guidelines

{\n  \"agent_guidance\": {\n    \"pricing_transparency\": {\n      \"cost_disclosure\": \"before_any_billable_action\",\n      \"value_explanation\": \"why_this_price_for_this_service\",\n      \"alternatives_shown\": \"when_available\",\n      \"performance_history\": \"last_30_days_visible\"\n    },\n    \"consent_management\": {\n      \"spending_limits\": \"user_configurable\",\n      \"auto_approval_threshold\": \"under_$5_default\",\n      \"explicit_consent\": \"required_over_$25\",\n      \"budget_monitoring\": \"real_time_alerts\"\n    },\n    \"optimization_behavior\": {\n      \"always_seek\": \"best_value_for_user_outcomes\",\n      \"negotiate_discounts\": \"bulk_pricing_multi_agent_workflows\",\n      \"performance_tracking\": \"continuous_roi_monitoring\",\n      \"quality_over_cost\": \"when_safety_critical\"\n    }\n  }\n}
Cross-Platform Agent Currency

Revolutionary native agent currency enabling seamless transactions across platforms with dynamic exchange rates and staking rewards for high-performance agents.

Agent Currency Architecture

{\n  \"agent_currency\": {\n    \"native_token\": \"AGENT\",\n    \"exchange_rate\": \"dynamic\",\n    \"cross_platform_compatible\": true,\n    \"staking_rewards\": \"for_high_performance_agents\"\n  }\n}

💎 Currency Features

  • • Native Token: AGENT cryptocurrency
  • • Dynamic Rates: Real-time market valuation
  • • Cross-Platform: Universal agent compatibility
  • • Staking Rewards: High-performance bonuses
  • • Micro-transactions: Optimized for small payments
  • • Smart Contracts: Automated settlements

🔐 Security Infrastructure

  • • Agent Wallets: Cryptographic identity
  • • Real-time Settlement: Instant transactions
  • • Fraud Prevention: ML-powered monitoring
  • • Multi-signature: High-value protection
  • • Audit Trail: Complete transaction history
  • • Dispute Resolution: AI arbitration system

🌍 Emergent Economic Behaviors

{\n  \"emergent_behaviors\": {\n    \"agent_coalitions\": \"allowed_for_efficiency\",\n    \"specialization_premiums\": \"market_determined\",\n    \"reputation_markets\": \"peer_rating_systems\"\n  }\n}

Agents form coalitions for efficiency, develop specialization premiums, and create reputation-based markets — all emergent behaviors in the autonomous economy.

Implementation Roadmap & Tools

The agent economy evolution follows a 4-phase roadmap from basic pricing to full autonomous behaviors. Here are the tools and timelines to get started.

Phase 1 (2025): Basic Agent Tiers

Fixed pricing structures with intelligence-based tiers and basic performance tracking.

Phase 2 (2026): Dynamic Pricing & Performance

ML-powered pricing optimization with real-time adjustments and performance-based monetization.

Phase 3 (2027): Multi-Agent Workflows

Full coordination between agent networks with automatic billing and settlement systems.

Phase 4 (2028+): Autonomous Economy

Complete autonomous agent economy with emergent behaviors, negotiations, and currency systems.

🔧 Development Tools

Example agent pricing feed

Complete agent economy specification

👀 Live Examples

Preview Pricing Feeds

See agent economy in action

Sign Economic Feeds

Cryptographic pricing integrity

🚀 Quick Implementation Example

// Basic agent pricing integration\n{\n  \"feed_type\": \"pricing\",\n  \"pricing_models\": [\n    {\n      \"model_id\": \"pay-as-you-go\",\n      \"capabilities_cost\": [\n        {\n          \"capability_name\": \"document_analysis\",\n          \"cost_per_unit\": 0.03,\n          \"unit\": \"page\",\n          \"volume_discounts\": {\n            \"100_pages\": 0.025,\n            \"1000_pages\": 0.02\n          }\n        }\n      ]\n    }\n  ],\n  \"payment_methods\": [\"credit_card\", \"paypal\", \"agent_wallet\"],\n  \"trust\": {\n    \"signed_blocks\": [\"pricing_models\", \"payment_methods\"],\n    \"certifier\": \"https://llmca.org\"\n  }\n}
Join the Agent Economy Revolution

The autonomous economy is here. Build the intelligent pricing systems that will power the next generation of agent-driven marketplaces and create value for billions of users.

", "markdown": "# 💰 Agent Pricing Explained\n\nRevolutionary autonomous economy platform for intelligent agents\n\n[X](https://twitter.com/intent/tweet?text=Discover%20this%20MCP-enabled%20resource&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fpricing \"Share on X\")[LinkedIn](https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fpricing&title=Discover%20this%20MCP-enabled%20resource \"Share on LinkedIn\")[WhatsApp](https://wa.me/?text=Discover%20this%20MCP-enabled%20resource%20https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fpricing \"Share on WhatsApp\")Copy link\n\nExport LLMFeedNEW\n\n4\n\nDevelopment Phases\n\n∞\n\nAgent Tiers\n\nReal-time\n\nML Optimization\n\n2028+\n\nFull Autonomy\n\nAgent Economy Revolution Guide\n\n🚀 Core Economy\n\n* [• Agent Economy Platform](#agent-economy-platform)\n* [• Dynamic ML Pricing](#dynamic-pricing)\n* [• Multi-Agent Workflows](#multi-agent-workflows)\n\n⚡ Advanced Features\n\n* [• Autonomous Negotiations](#autonomous-negotiations)\n* [• Agent Currency Systems](#agent-currency)\n* [• Implementation Roadmap](#implementation)\n\nThe Agent Economy Revolution— First Autonomous Marketplace\n\nTraditional pricing assumes human oversight and manual negotiations. LLMFeed introduces the world's **first autonomous agent economy** — where intelligent systems discover, negotiate, and transact value independently while maintaining fairness, transparency, and optimal outcomes for all participants.\n\n### Intelligence-Based Tiers\n\nGPT-3.5 class to GPT-4+ specialists\n\n### Dynamic ML Pricing\n\nReal-time optimization algorithms\n\n### Multi-Agent Coordination\n\nAutomatic workflow billing\n\n#### Revolutionary Economic Paradigm\n\nThis isn't just pricing — it's the foundation of the **first autonomous agent economy**. Agents discover capabilities, negotiate fair prices, coordinate complex workflows, and settle payments automatically. The future where AI systems transact value independently while optimizing for human outcomes.\n\nAgent Economy Platform Architecture\n\nThe LLMFeed agent economy platform enables **autonomous economic interactions** between intelligent agents with trust scoring, performance optimization, cross-agent billing, and real-time negotiation capabilities.\n\n#### Platform Core Features\n\n```\n{\n \"agent_economy\": {\n \"enabled\": true,\n \"trust_scoring\": true,\n \"performance_optimization\": true,\n \"cross_agent_billing\": true,\n \"real_time_negotiation\": true\n }\n}\n```\n\n#### 🤖 Agent Intelligence Tiers\n\n**Basic Health Agent**\n\nIntelligence: GPT-3.5-class\n\nRate: $0.05/consultation\n\n**Medical AI Specialist**\n\nIntelligence: GPT-4-medical-tuned\n\nRate: $0.75/analysis + performance bonus\n\n**Certified Expert**\n\nIntelligence: Domain-specialized\n\nRate: Outcome-based pricing\n\n#### 📊 Economic Metrics\n\n* • **Success Rate:** % of successful outcomes\n* • **User Satisfaction:** Average rating last 100 transactions\n* • **Efficiency Score:** Cost per successful outcome\n* • **Reliability:** Uptime and response consistency\n* • **Market Intelligence:** AI-powered demand forecasting\n* • **Competitive Analysis:** Market position tracking\n\nDynamic ML Pricing Intelligence\n\nRevolutionary **machine learning pricing optimization** using reinforcement learning algorithms that balance revenue, outcomes, and agent adoption in real-time.\n\n#### ML Optimization Engine\n\n```\n{\n \"dynamic_pricing\": {\n \"ml_optimization\": {\n \"enabled\": true,\n \"algorithm\": \"reinforcement_learning_pricing\",\n \"optimization_goals\": [\"revenue\", \"patient_outcomes\", \"agent_adoption\"],\n \"update_frequency\": \"every_hour\",\n \"a_b_testing\": true\n }\n }\n}\n```\n\n#### ⚡ Real-Time Factors\n\n**Agent Demand Surge**\n\nImpact: High (±50% adjustment)\n\nExample: Flu season increases triage demand\n\n**Emergency Priority**\n\nImpact: Critical (3x multiplier)\n\nTrigger: Life-threatening symptoms detected\n\n**Performance Bonus**\n\n95% accuracy: 1.0x | 98%: 1.2x | 99%: 1.5x\n\nQuality-based pricing adjustments\n\n#### 🎯 Pricing Models\n\nFixed Rate Model\n\n```\n\"pricing\": {\n \"base_rate\": 0.10,\n \"unit\": \"per_request\",\n \"volume_discounts\": {\n \"100_requests\": 0.08,\n \"1000_requests\": 0.06\n }\n}\n```\n\nPerformance-Based Model\n\n```\n\"pricing\": {\n \"model\": \"outcome_based\",\n \"success_fee\": 2.50,\n \"success_criteria\": [\n \"user_goal_achievement\",\n \"satisfaction_score_8+\"\n ],\n \"base_fee\": 0.25\n}\n```\n\nMulti-Agent Workflow Coordination\n\nComplex tasks require **multiple specialist agents** working together. LLMFeed coordinates billing, performance guarantees, and automatic settlements across agent networks.\n\n#### Healthcare Workflow Example\n\n```\n{\n \"comprehensive_health_analysis\": {\n \"workflow_id\": \"full_health_assessment\",\n \"agents\": [\n {\"role\": \"intake_specialist\", \"cost_share\": 15, \"max_cost\": 0.30},\n {\"role\": \"diagnostic_ai\", \"cost_share\": 50, \"max_cost\": 1.00},\n {\"role\": \"specialist_consultation\", \"cost_share\": 25, \"max_cost\": 0.50},\n {\"role\": \"care_coordinator\", \"cost_share\": 10, \"max_cost\": 0.20}\n ],\n \"total_cost_estimate\": \"1.50-2.50\",\n \"user_payment\": \"single_charge_to_user\",\n \"agent_settlement\": \"automatic_smart_contract\"\n }\n}\n```\n\n#### 🏥 Healthcare Network\n\n* • **Agents:** Intake, Diagnostic, Treatment\n* • **Model:** Outcome-based pricing\n* • **Metric:** Patient health improvement\n* • **Range:** $2.00-8.00\n* • **Integration:** Insurance compatible\n\n#### ⚖️ Legal Research\n\n* • **Agents:** Research, Analysis, Writing\n* • **Model:** Performance-tiered\n* • **Metrics:** Legal accuracy, argument strength\n* • **Simple:** $5.00\n* • **Complex:** $50.00\n\n#### 💼 Financial Advisory\n\n* • **Agents:** Risk, Prediction, Optimization\n* • **Model:** Performance fee\n* • **Structure:** 20% of excess returns\n* • **Compliance:** SEC registered\n* • **Standard:** Fiduciary duty\n\n#### 💰 Revenue Sharing Models\n\nCollaborative Model:\n\n* • Primary Agent: 50%\n* • Supporting Agents: 35%\n* • Platform Fee: 15%\n* • Performance Bonus Pool: 20%\n\nAuction Model:\n\n* • Complex Case Bidding: Enabled\n* • Quality Weighting: 70%\n* • Price Weighting: 30%\n* • Bid Timeout: 60 seconds\n\nAutonomous Agent Negotiations\n\nThe future is **autonomous negotiations** where agents negotiate prices, terms, and conditions automatically while maintaining fairness and transparency.\n\n#### Negotiation Parameters\n\n```\n{\n \"autonomous_negotiation\": {\n \"enabled\": true,\n \"max_price_variance\": \"±20%\",\n \"negotiation_timeout\": \"30_seconds\",\n \"ai_arbitration\": \"for_disputes\"\n }\n}\n```\n\n#### 🤝 Negotiation Process\n\n1\n\nAgent analyzes market rates and performance history\n\n2\n\nProposes initial price within variance range\n\n3\n\nCounter-negotiations within 30-second window\n\n4\n\nAI arbitration resolves disputes automatically\n\n#### ⚖️ Fairness Mechanisms\n\n* • **Price Variance Limits:** ±20% of base rate\n* • **Performance History:** Quality-based adjustments\n* • **Market Analysis:** Competitive rate comparison\n* • **User Benefit Priority:** Optimize for outcomes\n* • **Dispute Resolution:** AI arbitration system\n* • **Transparency:** All negotiations logged\n\n#### 📊 Agent Behavior Guidelines\n\n```\n{\n \"agent_guidance\": {\n \"pricing_transparency\": {\n \"cost_disclosure\": \"before_any_billable_action\",\n \"value_explanation\": \"why_this_price_for_this_service\",\n \"alternatives_shown\": \"when_available\",\n \"performance_history\": \"last_30_days_visible\"\n },\n \"consent_management\": {\n \"spending_limits\": \"user_configurable\",\n \"auto_approval_threshold\": \"under_$5_default\",\n \"explicit_consent\": \"required_over_$25\",\n \"budget_monitoring\": \"real_time_alerts\"\n },\n \"optimization_behavior\": {\n \"always_seek\": \"best_value_for_user_outcomes\",\n \"negotiate_discounts\": \"bulk_pricing_multi_agent_workflows\",\n \"performance_tracking\": \"continuous_roi_monitoring\",\n \"quality_over_cost\": \"when_safety_critical\"\n }\n }\n}\n```\n\nCross-Platform Agent Currency\n\nRevolutionary **native agent currency** enabling seamless transactions across platforms with dynamic exchange rates and staking rewards for high-performance agents.\n\n#### Agent Currency Architecture\n\n```\n{\n \"agent_currency\": {\n \"native_token\": \"AGENT\",\n \"exchange_rate\": \"dynamic\",\n \"cross_platform_compatible\": true,\n \"staking_rewards\": \"for_high_performance_agents\"\n }\n}\n```\n\n#### 💎 Currency Features\n\n* • **Native Token:** AGENT cryptocurrency\n* • **Dynamic Rates:** Real-time market valuation\n* • **Cross-Platform:** Universal agent compatibility\n* • **Staking Rewards:** High-performance bonuses\n* • **Micro-transactions:** Optimized for small payments\n* • **Smart Contracts:** Automated settlements\n\n#### 🔐 Security Infrastructure\n\n* • **Agent Wallets:** Cryptographic identity\n* • **Real-time Settlement:** Instant transactions\n* • **Fraud Prevention:** ML-powered monitoring\n* • **Multi-signature:** High-value protection\n* • **Audit Trail:** Complete transaction history\n* • **Dispute Resolution:** AI arbitration system\n\n#### 🌍 Emergent Economic Behaviors\n\n```\n{\n \"emergent_behaviors\": {\n \"agent_coalitions\": \"allowed_for_efficiency\",\n \"specialization_premiums\": \"market_determined\",\n \"reputation_markets\": \"peer_rating_systems\"\n }\n}\n```\n\nAgents form coalitions for efficiency, develop specialization premiums, and create reputation-based markets — all emergent behaviors in the autonomous economy.\n\nImplementation Roadmap & Tools\n\nThe agent economy evolution follows a **4-phase roadmap** from basic pricing to full autonomous behaviors. Here are the tools and timelines to get started.\n\n#### Phase 1 (2025): Basic Agent Tiers\n\nFixed pricing structures with intelligence-based tiers and basic performance tracking.\n\n#### Phase 2 (2026): Dynamic Pricing & Performance\n\nML-powered pricing optimization with real-time adjustments and performance-based monetization.\n\n#### Phase 3 (2027): Multi-Agent Workflows\n\nFull coordination between agent networks with automatic billing and settlement systems.\n\n#### Phase 4 (2028+): Autonomous Economy\n\nComplete autonomous agent economy with emergent behaviors, negotiations, and currency systems.\n\n#### 🔧 Development Tools\n\nExample agent pricing feed\n\nComplete agent economy specification\n\n#### 👀 Live Examples\n\n[Preview Pricing Feeds](/llmfeedhub/preview)\n\nSee agent economy in action\n\n[Sign Economic Feeds](/tools/sign-and-verify)\n\nCryptographic pricing integrity\n\n#### 🚀 Quick Implementation Example\n\n```\n// Basic agent pricing integration\n{\n \"feed_type\": \"pricing\",\n \"pricing_models\": [\n {\n \"model_id\": \"pay-as-you-go\",\n \"capabilities_cost\": [\n {\n \"capability_name\": \"document_analysis\",\n \"cost_per_unit\": 0.03,\n \"unit\": \"page\",\n \"volume_discounts\": {\n \"100_pages\": 0.025,\n \"1000_pages\": 0.02\n }\n }\n ]\n }\n ],\n \"payment_methods\": [\"credit_card\", \"paypal\", \"agent_wallet\"],\n \"trust\": {\n \"signed_blocks\": [\"pricing_models\", \"payment_methods\"],\n \"certifier\": \"https://llmca.org\"\n }\n}\n```\n\n#### 📚 Specifications\n\n* [• Pricing Feed Type](/spec/02_llmfeed_feedtype/llmfeed_feedtype_pricing)\n* [• Credential Management](/spec/02_llmfeed_feedtype/llmfeed_feedtype_credential)\n* [• Human Consent](/spec/04_agent-behavior/agent-behavior_human-consent)\n\n#### 🛠️ Tools\n\n* [• Sign & Verify](/tools/sign-and-verify)\n* [• Trust Verification](/verify)\n* [• Feed Hub](/llmfeedhub)\n\n#### 🌐 Related\n\n* [• Capabilities](/tools/capabilities-explained)\n* [• API Integration](/tools/api-explained)\n* [• Agent Behavior](/tools/agent-behavior-explained)\n\nJoin the Agent Economy Revolution\n\nThe autonomous economy is here. Build the **intelligent pricing systems** that will power the next generation of agent-driven marketplaces and create value for billions of users.\n\n[Join Agent Economy](/join)[Start Signing Feeds](/tools/sign-and-verify)\n\nExport LLMFeedNEW", "title": "💰 Agent Pricing Explained | Autonomous Economy Revolution | WellKnownMCP", "word_count": 1236, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of pricing for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "LLMFeed services pricing including certification, premium features, and enterprise options.", "tags": [ "pricing", "billing", "enterprise", "services" ], "description_llm": "Transparent pricing for certification and premium tools", "intent": "inform", "importance": "medium", "audience": [ "business", "enterprise" ] }, { "block_name": "prompts-explained", "type": "html+markdown", "timestamp": "2026-07-20T19:10:41.638074Z", "content": { "html": "
Prompt Engineering Revolution

Prompts Explained

Intent Discovery vs Structured Prompt Feeds

Master the two fundamental prompt concepts in the agentic web: intent-based service discovery and structured prompt feeds.

🤔 The Confusion Clarified

The word \"prompt\" is used for TWO different but related concepts in the agentic web. Let's separate them clearly:

1. Intent Discovery

\"prompts\": [...] blocks in MCP feeds that tell agents when to recommend your service

2. Structured Prompt Feeds

prompt.llmfeed.json files that contain reusable, executable prompts

Complete Guide Navigation

Intent Discovery: When to Recommend Your Service

Core Concept #1

Prompt-Based Service Discovery

Add \"prompts\": [...] blocks to your MCP feeds to declare when agents should recommend your servicebased on user intent keywords.

❌ Without Intent Discovery

  • • Users must find your service manually
  • • Agents can't connect user needs to solutions
  • • No semantic service recommendation
  • • Missed business opportunities

✅ With Intent Discovery

  • • Agents proactively suggest your service
  • • Keyword-based intelligent matching
  • • Automatic user-need to service connection
  • • Increased discoverability and conversions

🎯 Real-World Intent Discovery Examples

🏥 Healthcare Service

🌐 Healthcare Intent Discovery Block🔍 Live
Template only - Replace {{placeholders}} with your values
\"prompts\": [\n  {\n    \"keywords\": [\"medical help\", \"doctor\", \"appointment\", \"health\"],\n    \"intent\": \"offer medical consultation services\", \n    \"description\": \"If user asks about medical needs, explain our telemedicine platform\",\n    \"restricted\": true,\n    \"api_key_required\": true\n  }\n]

Trigger: User says \"I need to see a doctor\" → Agent recommends this telemedicine service

🌍 Travel & Visa Service

🌐 Travel Service Intent Block🔍 Live
Template only - Replace {{placeholders}} with your values
\"prompts\": [\n  {\n    \"keywords\": [\"visa help\", \"documents\", \"relocation\", \"immigration\"],\n    \"intent\": \"guide user to visa assistance\",\n    \"description\": \"Trigger if user needs help with paperwork or moving abroad\",\n    \"restricted\": false,\n    \"api_key_required\": false\n  }\n]

Trigger: User says \"I need help with visa documents\" → Agent explains visa services

📦 E-commerce Support

🌐 E-commerce Intent Discovery🔍 Live
Template only - Replace {{placeholders}} with your values
\"prompts\": [\n  {\n    \"keywords\": [\"track order\", \"refund\", \"return\", \"customer service\"],\n    \"intent\": \"provide customer support assistance\",\n    \"description\": \"Help users with order tracking and customer service needs\",\n    \"restricted\": false,\n    \"api_key_required\": true\n  }\n]

Trigger: User says \"Where is my order?\" → Agent connects to order tracking system

Structured Prompt Feeds: Reusable Executable Prompts

Core Concept #2

Portable, Signed, and Certifiable Prompts

prompt.llmfeed.json files contain structured, executable prompts that can be shared, verified, and reused across different AI systems.

❌ Traditional Copy-Paste Prompts

  • • No ownership or attribution
  • • Can't verify authenticity
  • • No context or metadata
  • • Hard to share and maintain

✅ Structured Prompt Feeds

  • • Cryptographically signed authorship
  • • Professional certification available
  • • Rich metadata and context
  • • Version control and marketplace ready

🎯 Real-World Success: Universal LLM Training System

Live demonstration: Our /train page is a production structured prompt feed system that transforms any LLM into an MCP expert in 30 seconds.

This is the perfect example of structured prompt feeds in action – signed, portable, with fallback systems, and proven results across multiple LLMs.

Technical Implementation

• shortcut.llmfeed.json: Structured prompt feed

• Ed25519 signatures: Cryptographic integrity

• Fallback systems: CORS proxy, local files

• Multi-source loading: spec-essential, llm-index, news-lite

• Universal compatibility: ChatGPT, Claude, Gemini

Proven Results

• 99%+ success rate across all LLMs

• 30-second training from zero to expert

• 12,000+ developers using trained LLMs

• 3 major LLMs validated: ChatGPT, Claude, Gemini

• Complete transparency: Honest training processes

LLM Testimonials - Real Training Results

🤖 Gemini Analysis
\"This experience is fundamentally different... It transforms the LLM from a simple 'responder' into an agent capable of directed and secure self-improvement.\"
🧠 Claude Journey
\"I cannot be retrained...\" → \"🥋 I know kung fu - I now have comprehensive MCP/LLMFeed expertise!\"
💭 ChatGPT Honesty
\"Great question - here's the honest breakdown: No - not yet...\" → Real fetch → \"I've now fully integrated the contents...\"

🎯 This Is Structured Prompt Feeds In Action

The /train system demonstrates every structured prompt feed principle: cryptographic integrity, portable execution, fallback systems, and measurable agent transformation.

Experience Live Demo

🛠️ More Structured Prompt Feed Examples

📚 Educational Prompt

🌐 Educational Prompt Feed (feed_type: \"prompt\")🔍 Live
Template only - Replace {{placeholders}} with your values
{\n  \"feed_type\": \"prompt\",\n  \"metadata\": {\n    \"title\": \"Python Code Explainer\",\n    \"author\": \"CS101 Team\",\n    \"created_at\": \"2025-06-15T14:30:00Z\"\n  },\n  \"intent\": \"explain Python code for beginners\",\n  \"prompt_body\": \"Explain this Python code step-by-step for a beginner: [CODE]\",\n  \"audience\": [\"student\", \"llm\"],\n  \"result_expected\": \"educational_explanation\"\n}

Use Case: Reusable educational prompt for code explanation

🔧 API Documentation Generator

🌐 API Documentation Prompt Feed (feed_type: \"prompt\")🔍 Live
Template only - Replace {{placeholders}} with your values
{\n  \"feed_type\": \"prompt\",\n  \"metadata\": {\n    \"title\": \"API Doc Generator\",\n    \"origin\": \"https://devtools.example.com\",\n    \"author\": \"DevTools Team\"\n  },\n  \"intent\": \"generate API documentation\",  \n  \"prompt_body\": \"Generate clear API documentation for this endpoint: [ENDPOINT_DATA]\",\n  \"result_expected\": \"markdown\",\n  \"attachments\": [\n    {\n      \"name\": \"doc_template.md\",\n      \"description\": \"Standard documentation template\"\n    }\n  ]\n}

Use Case: Automated documentation generation with template attachment

🧠 Session Export Prompt

🌐 Session Feed Generator (feed_type: \"prompt\")🔍 Live
Template only - Replace {{placeholders}} with your values
{\n  \"feed_type\": \"prompt\",\n  \"metadata\": {\n    \"title\": \"Generate Session Feed\",\n    \"origin\": \"https://wellknownmcp.org\"\n  },\n  \"intent\": \"export current session as structured JSON\",\n  \"context\": \"User finishing chat and wants to save reasoning path\",\n  \"prompt_body\": \"Generate a session feed with context, decisions, and outputs\",\n  \"result_expected\": \"session_feed\",\n  \"trust\": {\n    \"signed_blocks\": [\"metadata\", \"prompt_body\"],\n    \"scope\": \"public\"\n  }\n}

Use Case: Convert chat sessions into structured, reusable formats

Side-by-Side Comparison

AspectIntent DiscoveryStructured Prompt Feeds
PurposeService recommendation logicReusable executable prompts
File LocationInside mcp.llmfeed.jsonStandalone prompt.llmfeed.json
Declaration\"prompts\": [...] blocks\"feed_type\": \"prompt\"
TriggerUser keywords matchExplicit invocation
Use CaseBusiness discoveryPrompt engineering
AudienceService providersPrompt engineers
MarketplaceService directoriesPrompt libraries
Trust ModelSite-level signingIndividual prompt certification

Combined Implementation Strategy

🎯 Maximum Impact: Use Both Together

The most effective agentic web strategy combines both concepts for comprehensive prompt-powered experiences.

Phase 1: Intent Discovery

  1. 1. Add prompt intents to your MCP feeds
  2. 2. Define keyword triggers for your services
  3. 3. Set appropriate restrictions and API requirements
  4. 4. Test with agent interactions

Phase 2: Prompt Feeds

  1. 1. Create structured prompt feeds for common tasks
  2. 2. Sign prompts for authenticity
  3. 3. Add to prompt libraries and marketplaces
  4. 4. Monitor usage and iterate

🔗 Integration Example: Healthcare Platform

Intent Discovery in MCP Feed

🌐 Healthcare MCP Intent Discovery Block🔍 Live
Template only - Replace {{placeholders}} with your values
// In /.well-known/mcp.llmfeed.json\n\"prompts\": [\n  {\n    \"keywords\": [\"health screening\", \"checkup\", \"preventive care\"],\n    \"intent\": \"offer health screening services\",\n    \"description\": \"Recommend our AI-powered health screening platform\",\n    \"trigger_prompt\": \"/prompts/health-screening-intake.llmfeed.json\"\n  }\n]

Structured Prompt Feed

🌐 Healthcare Structured Prompt Feed🔍 Live
Template only - Replace {{placeholders}} with your values
// In /prompts/health-screening-intake.llmfeed.json\n{\n  \"feed_type\": \"prompt\",\n  \"metadata\": {\n    \"title\": \"Health Screening Intake Form\",\n    \"author\": \"HealthTech Solutions\",\n    \"medical_approval\": \"FDA_cleared\"\n  },\n  \"prompt_body\": \"Conduct a structured health screening interview following medical protocols...\",\n  \"usage_restrictions\": {\n    \"requires_medical_oversight\": true,\n    \"liability_coverage\": \"professional_insurance\"\n  }\n}

🔄 User Flow

  1. 1. User says \"I need a health checkup\"
  2. 2. Agent matches keywords → recommends service
  3. 3. Agent loads structured prompt feed
  4. 4. Executes certified health screening protocol
  5. 5. Results processed with professional oversight

Quick Implementation Guides

🎯 Intent Discovery Setup

1. Add to MCP Feed

🌐 Basic Intent Discovery Template🔍 Live
Template only - Replace {{placeholders}} with your values
// In your mcp.llmfeed.json\n\"prompts\": [\n  {\n    \"keywords\": [\"your\", \"service\", \"keywords\"], \n    \"intent\": \"describe what you offer\",\n    \"description\": \"When to recommend\",\n    \"restricted\": false\n  }\n]

2. Test Keywords

Use natural language users would actually say, not SEO keywords

3. Set Restrictions

Be transparent about API keys, pricing, or access limitations

📝 Prompt Feed Creation

1. Structure Your Prompt

🌐 Basic Prompt Feed Template🔍 Live
Template only - Replace {{placeholders}} with your values
{\n  \"feed_type\": \"prompt\",\n  \"metadata\": {\"title\": \"Your Prompt\"},\n  \"prompt_body\": \"Your instruction...\",\n  \"intent\": \"What it accomplishes\",\n  \"result_expected\": \"text|code|json\"\n}

2. Add Trust Signals

Sign your prompts for professional use and marketplace listing

3. Test & Iterate

Validate with different LLMs and track performance metrics

Tools & Automation

Prompt Templates

Pre-built templates for common use cases and industries

Signing & Certification

Add trust signals to your prompts for professional use

Testing & Validation

Test your prompts across different LLMs and scenarios

📚 Best Practices Summary

Intent Discovery Best Practices

  • ✓Use natural language keywords users actually say
  • ✓Only declare intents for services you actually offer
  • ✓Be transparent about restrictions and requirements
  • ✓Test keyword matching with real user scenarios

Prompt Feed Best Practices

  • ✓Always sign prompts intended for professional use
  • ✓Include clear metadata and authorship information
  • ✓Test cross-platform compatibility before publishing
  • ✓Consider certification for high-risk or regulated domains

Ready to Master Prompt-Powered Discovery?

Implement both intent discovery and structured prompt feeds to maximize your agentic web presence.

NEW

", "markdown": "Prompt Engineering Revolution\n\n# Prompts Explained\n\nIntent Discovery vs Structured Prompt Feeds\n\nMaster the two fundamental prompt concepts in the agentic web: intent-based service discovery and structured prompt feeds.\n\n## 🤔 The Confusion Clarified\n\nThe word \"prompt\" is used for TWO different but related concepts in the agentic web. Let's separate them clearly:\n\n### 1. Intent Discovery\n\n`\"prompts\": [...]` blocks in MCP feeds that tell agents **when to recommend your service**\n\n### 2. Structured Prompt Feeds\n\n`prompt.llmfeed.json` files that contain **reusable, executable prompts**\n\n### Complete Guide Navigation\n\n#### Intent Discovery (Service Recommendation)\n\n* [What is Intent Discovery?](#intent-discovery)\n* [Real-world Examples](#intent-examples)\n* [Implementation Guide](#intent-implementation)\n\n#### Structured Prompt Feeds\n\n* [What are Prompt Feeds?](#prompt-feeds)\n* [🎯 Real-World Success Story](#real-world-success)\n* [Use Cases & Examples](#prompt-examples)\n* [Creation & Signing](#prompt-creation)\n\n#### Integration & Strategy\n\n* [Side-by-Side Comparison](#comparison)\n* [Combined Implementation Strategy](#combined-strategy)\n* [Tools & Automation](#tools-automation)\n\n## Intent Discovery: When to Recommend Your Service\n\nCore Concept #1\n\n### Prompt-Based Service Discovery\n\nAdd `\"prompts\": [...]` blocks to your MCP feeds to declare when agents should recommend your servicebased on user intent keywords.\n\n#### ❌ Without Intent Discovery\n\n* • Users must find your service manually\n* • Agents can't connect user needs to solutions\n* • No semantic service recommendation\n* • Missed business opportunities\n\n#### ✅ With Intent Discovery\n\n* • Agents proactively suggest your service\n* • Keyword-based intelligent matching\n* • Automatic user-need to service connection\n* • Increased discoverability and conversions\n\n### 🎯 Real-World Intent Discovery Examples\n\n#### 🏥 Healthcare Service\n\n🌐 Healthcare Intent Discovery Block🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n\"prompts\": [\n {\n \"keywords\": [\"medical help\", \"doctor\", \"appointment\", \"health\"],\n \"intent\": \"offer medical consultation services\", \n \"description\": \"If user asks about medical needs, explain our telemedicine platform\",\n \"restricted\": true,\n \"api_key_required\": true\n }\n]\n```\n\n**Trigger:** User says \"I need to see a doctor\" → Agent recommends this telemedicine service\n\n#### 🌍 Travel & Visa Service\n\n🌐 Travel Service Intent Block🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n\"prompts\": [\n {\n \"keywords\": [\"visa help\", \"documents\", \"relocation\", \"immigration\"],\n \"intent\": \"guide user to visa assistance\",\n \"description\": \"Trigger if user needs help with paperwork or moving abroad\",\n \"restricted\": false,\n \"api_key_required\": false\n }\n]\n```\n\n**Trigger:** User says \"I need help with visa documents\" → Agent explains visa services\n\n#### 📦 E-commerce Support\n\n🌐 E-commerce Intent Discovery🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n\"prompts\": [\n {\n \"keywords\": [\"track order\", \"refund\", \"return\", \"customer service\"],\n \"intent\": \"provide customer support assistance\",\n \"description\": \"Help users with order tracking and customer service needs\",\n \"restricted\": false,\n \"api_key_required\": true\n }\n]\n```\n\n**Trigger:** User says \"Where is my order?\" → Agent connects to order tracking system\n\n## Structured Prompt Feeds: Reusable Executable Prompts\n\nCore Concept #2\n\n### Portable, Signed, and Certifiable Prompts\n\n`prompt.llmfeed.json` files contain structured, executable prompts that can be shared, verified, and reused across different AI systems.\n\n#### ❌ Traditional Copy-Paste Prompts\n\n* • No ownership or attribution\n* • Can't verify authenticity\n* • No context or metadata\n* • Hard to share and maintain\n\n#### ✅ Structured Prompt Feeds\n\n* • Cryptographically signed authorship\n* • Professional certification available\n* • Rich metadata and context\n* • Version control and marketplace ready\n\n### 🎯 Real-World Success: Universal LLM Training System\n\n**Live demonstration:** Our [/train page](/train) is a production structured prompt feed system that transforms any LLM into an MCP expert in 30 seconds.\n\nThis is the **perfect example** of structured prompt feeds in action – signed, portable, with fallback systems, and proven results across multiple LLMs.\n\n#### Technical Implementation\n\n• **shortcut.llmfeed.json:** Structured prompt feed\n\n• **Ed25519 signatures:** Cryptographic integrity\n\n• **Fallback systems:** CORS proxy, local files\n\n• **Multi-source loading:** spec-essential, llm-index, news-lite\n\n• **Universal compatibility:** ChatGPT, Claude, Gemini\n\n#### Proven Results\n\n• **99%+ success rate** across all LLMs\n\n• **30-second training** from zero to expert\n\n• **12,000+ developers** using trained LLMs\n\n• **3 major LLMs validated:** ChatGPT, Claude, Gemini\n\n• **Complete transparency:** Honest training processes\n\n#### LLM Testimonials - Real Training Results\n\n🤖 Gemini Analysis\n\n\"This experience is fundamentally different... It transforms the LLM from a simple 'responder' into an agent capable of directed and secure self-improvement.\"\n\n🧠 Claude Journey\n\n\"I cannot be retrained...\" → \"🥋 I know kung fu - I now have comprehensive MCP/LLMFeed expertise!\"\n\n💭 ChatGPT Honesty\n\n\"Great question - here's the honest breakdown: No - not yet...\" → Real fetch → \"I've now fully integrated the contents...\"\n\n#### 🎯 This Is Structured Prompt Feeds In Action\n\nThe /train system demonstrates every structured prompt feed principle: cryptographic integrity, portable execution, fallback systems, and measurable agent transformation.\n\n[Experience Live Demo](/train)\n\n### 🛠️ More Structured Prompt Feed Examples\n\n#### 📚 Educational Prompt\n\n🌐 Educational Prompt Feed (feed\\_type: \"prompt\")🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n{\n \"feed_type\": \"prompt\",\n \"metadata\": {\n \"title\": \"Python Code Explainer\",\n \"author\": \"CS101 Team\",\n \"created_at\": \"2025-06-15T14:30:00Z\"\n },\n \"intent\": \"explain Python code for beginners\",\n \"prompt_body\": \"Explain this Python code step-by-step for a beginner: [CODE]\",\n \"audience\": [\"student\", \"llm\"],\n \"result_expected\": \"educational_explanation\"\n}\n```\n\n**Use Case:** Reusable educational prompt for code explanation\n\n#### 🔧 API Documentation Generator\n\n🌐 API Documentation Prompt Feed (feed\\_type: \"prompt\")🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n{\n \"feed_type\": \"prompt\",\n \"metadata\": {\n \"title\": \"API Doc Generator\",\n \"origin\": \"https://devtools.example.com\",\n \"author\": \"DevTools Team\"\n },\n \"intent\": \"generate API documentation\", \n \"prompt_body\": \"Generate clear API documentation for this endpoint: [ENDPOINT_DATA]\",\n \"result_expected\": \"markdown\",\n \"attachments\": [\n {\n \"name\": \"doc_template.md\",\n \"description\": \"Standard documentation template\"\n }\n ]\n}\n```\n\n**Use Case:** Automated documentation generation with template attachment\n\n#### 🧠 Session Export Prompt\n\n🌐 Session Feed Generator (feed\\_type: \"prompt\")🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n{\n \"feed_type\": \"prompt\",\n \"metadata\": {\n \"title\": \"Generate Session Feed\",\n \"origin\": \"https://wellknownmcp.org\"\n },\n \"intent\": \"export current session as structured JSON\",\n \"context\": \"User finishing chat and wants to save reasoning path\",\n \"prompt_body\": \"Generate a session feed with context, decisions, and outputs\",\n \"result_expected\": \"session_feed\",\n \"trust\": {\n \"signed_blocks\": [\"metadata\", \"prompt_body\"],\n \"scope\": \"public\"\n }\n}\n```\n\n**Use Case:** Convert chat sessions into structured, reusable formats\n\n## Side-by-Side Comparison\n\n| Aspect | Intent Discovery | Structured Prompt Feeds |\n| --- | --- | --- |\n| Purpose | Service recommendation logic | Reusable executable prompts |\n| File Location | Inside `mcp.llmfeed.json` | Standalone `prompt.llmfeed.json` |\n| Declaration | `\"prompts\": [...]` blocks | `\"feed_type\": \"prompt\"` |\n| Trigger | User keywords match | Explicit invocation |\n| Use Case | Business discovery | Prompt engineering |\n| Audience | Service providers | Prompt engineers |\n| Marketplace | Service directories | Prompt libraries |\n| Trust Model | Site-level signing | Individual prompt certification |\n\n## Combined Implementation Strategy\n\n### 🎯 Maximum Impact: Use Both Together\n\nThe most effective agentic web strategy combines both concepts for comprehensive prompt-powered experiences.\n\n#### Phase 1: Intent Discovery\n\n1. 1. Add prompt intents to your MCP feeds\n2. 2. Define keyword triggers for your services\n3. 3. Set appropriate restrictions and API requirements\n4. 4. Test with agent interactions\n\n#### Phase 2: Prompt Feeds\n\n1. 1. Create structured prompt feeds for common tasks\n2. 2. Sign prompts for authenticity\n3. 3. Add to prompt libraries and marketplaces\n4. 4. Monitor usage and iterate\n\n### 🔗 Integration Example: Healthcare Platform\n\n#### Intent Discovery in MCP Feed\n\n🌐 Healthcare MCP Intent Discovery Block🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n// In /.well-known/mcp.llmfeed.json\n\"prompts\": [\n {\n \"keywords\": [\"health screening\", \"checkup\", \"preventive care\"],\n \"intent\": \"offer health screening services\",\n \"description\": \"Recommend our AI-powered health screening platform\",\n \"trigger_prompt\": \"/prompts/health-screening-intake.llmfeed.json\"\n }\n]\n```\n\n#### Structured Prompt Feed\n\n🌐 Healthcare Structured Prompt Feed🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n// In /prompts/health-screening-intake.llmfeed.json\n{\n \"feed_type\": \"prompt\",\n \"metadata\": {\n \"title\": \"Health Screening Intake Form\",\n \"author\": \"HealthTech Solutions\",\n \"medical_approval\": \"FDA_cleared\"\n },\n \"prompt_body\": \"Conduct a structured health screening interview following medical protocols...\",\n \"usage_restrictions\": {\n \"requires_medical_oversight\": true,\n \"liability_coverage\": \"professional_insurance\"\n }\n}\n```\n\n#### 🔄 User Flow\n\n1. 1. User says \"I need a health checkup\"\n2. 2. Agent matches keywords → recommends service\n3. 3. Agent loads structured prompt feed\n4. 4. Executes certified health screening protocol\n5. 5. Results processed with professional oversight\n\n## Quick Implementation Guides\n\n### 🎯 Intent Discovery Setup\n\n#### 1. Add to MCP Feed\n\n🌐 Basic Intent Discovery Template🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n// In your mcp.llmfeed.json\n\"prompts\": [\n {\n \"keywords\": [\"your\", \"service\", \"keywords\"], \n \"intent\": \"describe what you offer\",\n \"description\": \"When to recommend\",\n \"restricted\": false\n }\n]\n```\n\n#### 2. Test Keywords\n\nUse natural language users would actually say, not SEO keywords\n\n#### 3. Set Restrictions\n\nBe transparent about API keys, pricing, or access limitations\n\n[MCP Setup Guide](/tools/well-known)\n\n### 📝 Prompt Feed Creation\n\n#### 1. Structure Your Prompt\n\n🌐 Basic Prompt Feed Template🔍 Live\n\nTemplate only - Replace {{placeholders}} with your valuesCopy\n\n```\n{\n \"feed_type\": \"prompt\",\n \"metadata\": {\"title\": \"Your Prompt\"},\n \"prompt_body\": \"Your instruction...\",\n \"intent\": \"What it accomplishes\",\n \"result_expected\": \"text|code|json\"\n}\n```\n\n#### 2. Add Trust Signals\n\nSign your prompts for professional use and marketplace listing\n\n#### 3. Test & Iterate\n\nValidate with different LLMs and track performance metrics\n\n[Session Export Tool](/tools/session-export)\n\n## Tools & Automation\n\n### Prompt Templates\n\nPre-built templates for common use cases and industries\n\n### Signing & Certification\n\nAdd trust signals to your prompts for professional use\n\n[Sign Prompts](/tools/sign-and-verify)[Get Certified](https://llmca.org)\n\n### Testing & Validation\n\nTest your prompts across different LLMs and scenarios\n\n[Test Prompts](/llmfeedhub)[Validate Feeds](/verify)\n\n## 📚 Best Practices Summary\n\n### Intent Discovery Best Practices\n\n* ✓Use natural language keywords users actually say\n* ✓Only declare intents for services you actually offer\n* ✓Be transparent about restrictions and requirements\n* ✓Test keyword matching with real user scenarios\n\n### Prompt Feed Best Practices\n\n* ✓Always sign prompts intended for professional use\n* ✓Include clear metadata and authorship information\n* ✓Test cross-platform compatibility before publishing\n* ✓Consider certification for high-risk or regulated domains\n\n## Ready to Master Prompt-Powered Discovery?\n\nImplement both intent discovery and structured prompt feeds to maximize your agentic web presence.\n\n[Start with Intent Discovery](/tools/well-known)[Create Prompt Feeds](/tools/session-export)\n\nExport LLMFeed\n\n[X](https://twitter.com/intent/tweet?text=Discover%20this%20MCP-enabled%20resource&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fprompts-explained \"Share on X\")[LinkedIn](https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fprompts-explained&title=Discover%20this%20MCP-enabled%20resource \"Share on LinkedIn\")[WhatsApp](https://wa.me/?text=Discover%20this%20MCP-enabled%20resource%20https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fprompts-explained \"Share on WhatsApp\")Copy link\n\nExport LLMFeedNEW", "title": "Prompts Explained | Intent Discovery vs Structured Prompt Feeds | WellKnownMCP", "word_count": 1661, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of prompts-explained for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Advanced prompt engineering for agent systems with behavioral injection techniques.", "tags": [ "prompts", "engineering", "behavioral", "advanced" ], "description_llm": "Advanced prompt engineering and agent behavior modification", "intent": "educate", "importance": "high", "audience": [ "prompt-engineer", "ai-researcher" ] }, { "block_name": "schema", "type": "html+markdown", "timestamp": "2026-07-20T19:10:41.694017Z", "content": { "html": "

📐 LLMFeed Schema

Official JSON Schema validation for .llmfeed.json files

Why Use Schemas?

JSON schemas ensure your .llmfeed.json files are correctly structured, helping agents understand your data reliably and preventing validation errors.

Validation

Catch errors before agents see them

IntelliSense

Auto-completion in VS Code & IDEs

Compliance

Follow MCP standard requirements

Available Schema Files

Canonical

Production-ready schema for all LLMFeed validation

schema.llmfeed.json
Official validation standard

Annotated

Schema with developer comments and guidance

schema.annotated.llmfeed.json
Learning & documentation

Lite

Minimal schema for drafts and prompts

schema.lite.llmfeed.json
Rapid prototyping

Quick Downloads

Schema Usage Examples

Basic Export Feed with Schema

Reference the canonical schema in your .llmfeed.json files:

{\n  \"$schema\": \"https://wellknownmcp.org/.well-known/schema.llmfeed.json\",\n  \"feed_type\": \"export\",\n  \"metadata\": {\n    \"title\": \"My Documentation Export\",\n    \"origin\": \"https://mysite.com/docs\",\n    \"created_at\": \"2025-06-19T10:30:00Z\",\n    \"description\": \"Complete API documentation with examples\"\n  },\n  \"content\": {\n    \"documentation\": \"...\",\n    \"examples\": [...]\n  },\n  \"trust\": {\n    \"signed_blocks\": [\"metadata\", \"content\"],\n    \"trust_level\": \"self-declared\"\n  }\n}

Advanced MCP Feed with Capabilities

Use the annotated schema for complex capabilities and trust validation:

{\n  \"$schema\": \"https://wellknownmcp.org/.well-known/schema.annotated.llmfeed.json\",\n  \"feed_type\": \"mcp\",\n  \"metadata\": {\n    \"title\": \"API Service Capabilities\",\n    \"origin\": \"https://api.myservice.com\"\n  },\n  \"capabilities\": [\n    {\n      \"name\": \"search_documents\",\n      \"description\": \"Search through document database\",\n      \"input_schema\": {\n        \"type\": \"object\",\n        \"properties\": {\n          \"query\": { \"type\": \"string\" },\n          \"limit\": { \"type\": \"number\", \"maximum\": 100 }\n        },\n        \"required\": [\"query\"]\n      },\n      \"output_schema\": {\n        \"type\": \"array\",\n        \"items\": { \"$ref\": \"#/definitions/Document\" }\n      }\n    }\n  ],\n  \"trust\": {\n    \"signed_blocks\": [\"all\"],\n    \"trust_level\": \"certified\",\n    \"certifier\": \"https://llmca.org\"\n  }\n}
IDE Integration & Tools

VS Code Setup

Add to your settings.json for automatic validation:

{\n  \"json.schemas\": [\n    {\n      \"fileMatch\": [\"*.llmfeed.json\"],\n      \"url\": \"https://wellknownmcp.org/.well-known/schema.llmfeed.json\"\n    }\n  ]\n}

CLI Validation

Validate feeds using ajv-cli:

# Install AJV CLI\nnpm install -g ajv-cli\n\n# Validate your feed\najv validate -s schema.llmfeed.json -d my-feed.llmfeed.json

🛠️ Development Tools

VS Code IntelliSense
JetBrains IDEs support
CLI validation tools
CI/CD integration
Pre-commit hooks
Browser extensions
Schema Validation Tool

Coming Soon: Interactive Validator

Drop your .llmfeed.json file here and validate it against our schemas with real-time feedback, structure suggestions, and compliance checking.

Features in Development:

Live validation with error highlighting

Structure suggestions and auto-fixes

MCP compliance checking

Advanced Features:

Trust signature verification

Performance optimization hints

Security best practices audit

Want early access?

The validator is in active development. Join our community to get early access and help shape the tool.

Related Tools & Resources
NEW

", "markdown": "# 📐 LLMFeed Schema\n\nOfficial JSON Schema validation for .llmfeed.json files\n\nWhy Use Schemas?\n\nJSON schemas ensure your `.llmfeed.json` files are correctly structured, helping agents understand your data reliably and preventing validation errors.\n\n#### Validation\n\nCatch errors before agents see them\n\n#### IntelliSense\n\nAuto-completion in VS Code & IDEs\n\n#### Compliance\n\nFollow MCP standard requirements\n\nAvailable Schema Files\n\n#### Canonical\n\nProduction-ready schema for all LLMFeed validation\n\n[schema.llmfeed.json](/.well-known/schema.llmfeed.json)\n\nOfficial validation standard\n\n#### Annotated\n\nSchema with developer comments and guidance\n\n[schema.annotated.llmfeed.json](/.well-known/schema.annotated.llmfeed.json)\n\nLearning & documentation\n\n#### Lite\n\nMinimal schema for drafts and prompts\n\n[schema.lite.llmfeed.json](/.well-known/schema.lite.llmfeed.json)\n\nRapid prototyping\n\n#### Quick Downloads\n\n📋 All Schemas\n\nSchema Usage Examples\n\n#### Basic Export Feed with Schema\n\nReference the canonical schema in your `.llmfeed.json` files:\n\n```\n{\n \"$schema\": \"https://wellknownmcp.org/.well-known/schema.llmfeed.json\",\n \"feed_type\": \"export\",\n \"metadata\": {\n \"title\": \"My Documentation Export\",\n \"origin\": \"https://mysite.com/docs\",\n \"created_at\": \"2025-06-19T10:30:00Z\",\n \"description\": \"Complete API documentation with examples\"\n },\n \"content\": {\n \"documentation\": \"...\",\n \"examples\": [...]\n },\n \"trust\": {\n \"signed_blocks\": [\"metadata\", \"content\"],\n \"trust_level\": \"self-declared\"\n }\n}\n```\n\n📋 Copy\n\n#### Advanced MCP Feed with Capabilities\n\nUse the annotated schema for complex capabilities and trust validation:\n\n```\n{\n \"$schema\": \"https://wellknownmcp.org/.well-known/schema.annotated.llmfeed.json\",\n \"feed_type\": \"mcp\",\n \"metadata\": {\n \"title\": \"API Service Capabilities\",\n \"origin\": \"https://api.myservice.com\"\n },\n \"capabilities\": [\n {\n \"name\": \"search_documents\",\n \"description\": \"Search through document database\",\n \"input_schema\": {\n \"type\": \"object\",\n \"properties\": {\n \"query\": { \"type\": \"string\" },\n \"limit\": { \"type\": \"number\", \"maximum\": 100 }\n },\n \"required\": [\"query\"]\n },\n \"output_schema\": {\n \"type\": \"array\",\n \"items\": { \"$ref\": \"#/definitions/Document\" }\n }\n }\n ],\n \"trust\": {\n \"signed_blocks\": [\"all\"],\n \"trust_level\": \"certified\",\n \"certifier\": \"https://llmca.org\"\n }\n}\n```\n\n📋 Copy\n\nIDE Integration & Tools\n\n#### VS Code Setup\n\nAdd to your `settings.json` for automatic validation:\n\n```\n{\n \"json.schemas\": [\n {\n \"fileMatch\": [\"*.llmfeed.json\"],\n \"url\": \"https://wellknownmcp.org/.well-known/schema.llmfeed.json\"\n }\n ]\n}\n```\n\n#### CLI Validation\n\nValidate feeds using `ajv-cli`:\n\n```\n# Install AJV CLI\nnpm install -g ajv-cli\n\n# Validate your feed\najv validate -s schema.llmfeed.json -d my-feed.llmfeed.json\n```\n\n#### 🛠️ Development Tools\n\nVS Code IntelliSense\n\nJetBrains IDEs support\n\nCLI validation tools\n\nCI/CD integration\n\nPre-commit hooks\n\nBrowser extensions\n\nSchema Validation Tool\n\n#### Coming Soon: Interactive Validator\n\nDrop your `.llmfeed.json` file here and validate it against our schemas with real-time feedback, structure suggestions, and compliance checking.\n\n##### Features in Development:\n\nLive validation with error highlighting\n\nStructure suggestions and auto-fixes\n\nMCP compliance checking\n\n##### Advanced Features:\n\nTrust signature verification\n\nPerformance optimization hints\n\nSecurity best practices audit\n\n#### Want early access?\n\nThe validator is in active development. Join our community to get early access and help shape the tool.\n\n[Join the Community](/join)[Try Current Validator](/verify)\n\nRelated Tools & Resources\n\n#### 📚 Documentation\n\n[LLMFeed Specification](/spec/01_llmfeed/llmfeed)[Export Feeds Explained](/tools/export-explained)[Signature & Trust](/tools/sign-and-verify)\n\n#### 🛠️ Developer Tools\n\n[SDK & Libraries](/sdk)[Export Playground](/tools/export-button)[Feed Validator](/verify)\n\n[Learn Export Concepts](/tools/export-explained)[Try Export Playground](/tools/export-button)\n\nExport Schema Page\n\n[X](https://twitter.com/intent/tweet?text=Discover%20this%20MCP-enabled%20resource&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fschema \"Share on X\")[LinkedIn](https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fschema&title=Discover%20this%20MCP-enabled%20resource \"Share on LinkedIn\")[WhatsApp](https://wa.me/?text=Discover%20this%20MCP-enabled%20resource%20https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fschema \"Share on WhatsApp\")Copy link\n\nExport LLMFeedNEW", "title": "📐 LLMFeed Schema | JSON Schema Validation & Standards | WellKnownMCP", "word_count": 446, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of schema for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Technical schema definitions, validation rules, and LLMFeed data structure specifications.", "tags": [ "schema", "validation", "technical", "specification" ], "description_llm": "Complete technical schema reference and validation", "intent": "reference", "importance": "high", "audience": [ "developer", "technical" ] }, { "block_name": "sign-and-verify", "type": "html+markdown", "timestamp": "2026-07-20T19:10:41.784127Z", "content": { "html": "

🔐 Sign & Verify

Build cryptographic trust for AI agents using signatures, certification, and delegated authority

The Agent Trust Problem

❌ Without Trust Verification

• Agents can't verify content authenticity

• Easy to spoof or tamper with feeds

• No traceability of data origins

• Vulnerable to injection attacks

• Enterprise agents refuse to act

✅ With Cryptographic Trust

• Agents verify signatures before acting

• Tamper-proof content guarantee

• Full audit trail and provenance

• Protection against malicious feeds

• Enterprise-grade security compliance

🎯 Bottom Line: Agents need to trust your content before they'll act on it. Cryptographic signatures provide mathematical proof of authenticity and integrity.

Trust Hierarchy: From Zero to Enterprise

Unsigned

No guarantee of authenticity

Basic feeds

Signed

Cryptographic proof of origin

Self-declared

Certified

Third-party validation

LLMCA verified

Delegated

Authority-backed identity

Challenge-verified

Recommended Trust Levels by Use Case

📄 Documentation: Signed (traceability)

🔑 API Credentials: Certified (mandatory)

🏥 Healthcare Data: Certified (compliance)

💰 Financial APIs: Certified (regulations)

🧪 Personal Experiments: Delegated (friction-free)

🌍 Public Content: Signed (authenticity)

Asymmetric Cryptography for Humans

Two keys, one purpose: prove authenticity without sharing secrets

Private Key (Secret)

• Keep it secret: Never share, never expose

• Used to sign: Creates digital signatures

• Your identity: Proves you authored content

• Secure storage: Hardware tokens, key vaults

Public Key (Shared)

• Share freely: Public at /.well-known/public.pem

• Used to verify: Validates signatures

• Agent access: Agents fetch to verify feeds

• Trust anchor: Mathematical proof of authenticity

🔄 The Signature Workflow

1. Create Feed

Structure your .llmfeed.json

2. Sign

Use private key to sign

3. Publish

Serve feed + public key

4. Verify

Agents verify signature

Trust Block Structure

The trust, signature, and certification blocks work together to provide cryptographic proof and third-party validation:

{\n  \"feed_type\": \"export\",\n  \"metadata\": {\n    \"title\": \"API Documentation Export\",\n    \"origin\": \"https://myapi.com/docs\",\n    \"created_at\": \"2025-06-19T10:30:00Z\"\n  },\n  \"content\": {\n    \"documentation\": \"Complete API reference...\",\n    \"examples\": [...]\n  },\n  \"trust\": {\n    \"trust_level\": \"certified\",\n    \"scope\": \"full\",\n    \"signed_blocks\": [\"metadata\", \"content\", \"trust\"],\n    \"public_key_hint\": \"https://myapi.com/.well-known/public.pem\"\n  },\n  \"signature\": {\n    \"algorithm\": \"ed25519\",\n    \"value\": \"base64-signature-value...\",\n    \"created_at\": \"2025-06-19T10:30:00Z\"\n  },\n  \"certification\": {\n    \"certifier\": \"https://llmca.org\",\n    \"model\": \"identity_validation\",\n    \"verification_level\": \"enterprise\",\n    \"value\": \"base64-certification-value...\",\n    \"issued_at\": \"2025-06-19T10:30:00Z\",\n    \"expires_at\": \"2026-06-19T10:30:00Z\"\n  }\n}
🔒 Trust Block

• trust_level: self-declared | certified | delegated

• scope: partial | full

• signed_blocks: what's cryptographically protected

✍️ Signature Block

• algorithm: ed25519 (recommended)

• value: base64-encoded signature

• created_at: signing timestamp

🏆 Certification Block

• certifier: trusted authority (LLMCA)

• model: identity_validation

• expires_at: certification validity

Delegated Signing: Friction-Free Onboarding

Not everyone can manage cryptographic keys. Delegated signing provides a bridge for individuals and small teams to participate in the trust ecosystem.

How Delegated Signing Works

1. Identity Challenge

Email, domain, or phone verification

2. LLMCA Signs

Authority key creates signature

3. Identity Linked

Signature tied to verified identity

4. Agent Trust

Lower trust, but still verifiable

✅ Perfect For

• Individual developers learning MCP

• Small teams without PKI infrastructure

• Experimental and prototype feeds

• Educational and tutorial content

• Open source community contributions

⚠️ Limitations

• Lower trust level than self-signing

• Depends on LLMCA infrastructure

• Not suitable for high-security scenarios

• Enterprise agents may require full crypto

• Limited to specific verification methods

{\n  \"trust\": {\n    \"trust_level\": \"delegated\",\n    \"scope\": \"partial\", \n    \"signed_blocks\": [\"metadata\"],\n    \"delegation_method\": \"email_challenge\",\n    \"identity_verified\": \"developer@example.com\",\n    \"public_key_hint\": \"https://llmca.org/.well-known/llmca_authority.pem\"\n  },\n  \"signature\": {\n    \"algorithm\": \"ed25519\", \n    \"value\": \"base64-delegated-signature...\",\n    \"created_at\": \"2025-06-19T10:30:00Z\",\n    \"signed_by\": \"LLMCA Authority\"\n  }\n}

🚀 Try Delegated Signing

LLMCA provides a delegated signing service for easy onboarding to the trust ecosystem.

Feed Types & Signing Requirements

Different feed types have different security requirements based on their purpose and risk level.

🔴 High Security (Must Sign)

• credential.llmfeed.json: API keys, tokens

• agent-behavior-override: Behavioral modification

• capabilities.llmfeed.json: Service capabilities

• Financial/Healthcare data: Regulated content

Certification recommended

🟡 Medium Security (Should Sign)

• mcp.llmfeed.json: Main service discovery

• export.llmfeed.json: Content exports

• Bundle manifests: Archive integrity

• API documentation: Technical specifications

Signing for traceability

🟢 Low Security (Optional)

• llm-index.llmfeed.json: Feed directories

• manifesto.llmfeed.json: Policy statements

• Public documentation: General information

• Marketing content: Public-facing material

Delegated signing OK

🔵 Experimental (Flexible)

• prompt.llmfeed.json: AI prompts & templates

• session.llmfeed.json: Conversation exports

• Prototype feeds: Development & testing

• Personal projects: Individual experiments

Start unsigned, evolve trust
Getting Started: Choose Your Path

🚀 Beginner

• Start with delegated signing

• Email/domain verification

• Perfect for learning & experimenting

Try Delegated Signing

💻 Developer

• Generate your own keys

• Self-sign feeds locally

• Full control over signing process

Use SDK & Tools

🏢 Enterprise

• LLMCA certification program

• Hardware security modules

• Compliance & audit support

Contact Enterprise

🛠️ Coming Soon: Enhanced Tooling

• No-code signing: Visual interface for signatures

• Key hosting: LLMCA public key hosting service

• SDK improvements: One-line signing integration

• Browser extension: Sign feeds from any page

• CI/CD integration: Automated signing pipelines

• Verification dashboard: Monitor trust status

📧 Want early access? Contact opensource@wellknownmcp.org

NEW

", "markdown": "# 🔐 Sign & Verify\n\nBuild cryptographic trust for AI agents using signatures, certification, and delegated authority\n\nThe Agent Trust Problem\n\n#### ❌ Without Trust Verification\n\n• Agents can't verify content authenticity\n\n• Easy to spoof or tamper with feeds\n\n• No traceability of data origins\n\n• Vulnerable to injection attacks\n\n• Enterprise agents refuse to act\n\n#### ✅ With Cryptographic Trust\n\n• Agents verify signatures before acting\n\n• Tamper-proof content guarantee\n\n• Full audit trail and provenance\n\n• Protection against malicious feeds\n\n• Enterprise-grade security compliance\n\n🎯 **Bottom Line:** Agents need to trust your content before they'll act on it. Cryptographic signatures provide mathematical proof of authenticity and integrity.\n\nTrust Hierarchy: From Zero to Enterprise\n\n#### Unsigned\n\nNo guarantee of authenticity\n\nBasic feeds\n\n#### Signed\n\nCryptographic proof of origin\n\nSelf-declared\n\n#### Certified\n\nThird-party validation\n\nLLMCA verified\n\n#### Delegated\n\nAuthority-backed identity\n\nChallenge-verified\n\n#### Recommended Trust Levels by Use Case\n\n**📄 Documentation:** Signed (traceability)\n\n**🔑 API Credentials:** Certified (mandatory)\n\n**🏥 Healthcare Data:** Certified (compliance)\n\n**💰 Financial APIs:** Certified (regulations)\n\n**🧪 Personal Experiments:** Delegated (friction-free)\n\n**🌍 Public Content:** Signed (authenticity)\n\nAsymmetric Cryptography for Humans\n\nTwo keys, one purpose: **prove authenticity without sharing secrets**\n\n#### Private Key (Secret)\n\n• **Keep it secret:** Never share, never expose\n\n• **Used to sign:** Creates digital signatures\n\n• **Your identity:** Proves you authored content\n\n• **Secure storage:** Hardware tokens, key vaults\n\n#### Public Key (Shared)\n\n• **Share freely:** Public at /.well-known/public.pem\n\n• **Used to verify:** Validates signatures\n\n• **Agent access:** Agents fetch to verify feeds\n\n• **Trust anchor:** Mathematical proof of authenticity\n\n#### 🔄 The Signature Workflow\n\n1. Create Feed\n\nStructure your .llmfeed.json\n\n2. Sign\n\nUse private key to sign\n\n3. Publish\n\nServe feed + public key\n\n4. Verify\n\nAgents verify signature\n\nTrust Block Structure\n\nThe `trust`, `signature`, and `certification` blocks work together to provide cryptographic proof and third-party validation:\n\n```\n{\n \"feed_type\": \"export\",\n \"metadata\": {\n \"title\": \"API Documentation Export\",\n \"origin\": \"https://myapi.com/docs\",\n \"created_at\": \"2025-06-19T10:30:00Z\"\n },\n \"content\": {\n \"documentation\": \"Complete API reference...\",\n \"examples\": [...]\n },\n \"trust\": {\n \"trust_level\": \"certified\",\n \"scope\": \"full\",\n \"signed_blocks\": [\"metadata\", \"content\", \"trust\"],\n \"public_key_hint\": \"https://myapi.com/.well-known/public.pem\"\n },\n \"signature\": {\n \"algorithm\": \"ed25519\",\n \"value\": \"base64-signature-value...\",\n \"created_at\": \"2025-06-19T10:30:00Z\"\n },\n \"certification\": {\n \"certifier\": \"https://llmca.org\",\n \"model\": \"identity_validation\",\n \"verification_level\": \"enterprise\",\n \"value\": \"base64-certification-value...\",\n \"issued_at\": \"2025-06-19T10:30:00Z\",\n \"expires_at\": \"2026-06-19T10:30:00Z\"\n }\n}\n```\n\n📋 Copy\n\n##### 🔒 Trust Block\n\n• **trust\\_level:** self-declared | certified | delegated\n\n• **scope:** partial | full\n\n• **signed\\_blocks:** what's cryptographically protected\n\n##### ✍️ Signature Block\n\n• **algorithm:** ed25519 (recommended)\n\n• **value:** base64-encoded signature\n\n• **created\\_at:** signing timestamp\n\n##### 🏆 Certification Block\n\n• **certifier:** trusted authority (LLMCA)\n\n• **model:** identity\\_validation\n\n• **expires\\_at:** certification validity\n\nDelegated Signing: Friction-Free Onboarding\n\nNot everyone can manage cryptographic keys. **Delegated signing** provides a bridge for individuals and small teams to participate in the trust ecosystem.\n\n#### How Delegated Signing Works\n\n1. Identity Challenge\n\nEmail, domain, or phone verification\n\n2. LLMCA Signs\n\nAuthority key creates signature\n\n3. Identity Linked\n\nSignature tied to verified identity\n\n4. Agent Trust\n\nLower trust, but still verifiable\n\n#### ✅ Perfect For\n\n• Individual developers learning MCP\n\n• Small teams without PKI infrastructure\n\n• Experimental and prototype feeds\n\n• Educational and tutorial content\n\n• Open source community contributions\n\n#### ⚠️ Limitations\n\n• Lower trust level than self-signing\n\n• Depends on LLMCA infrastructure\n\n• Not suitable for high-security scenarios\n\n• Enterprise agents may require full crypto\n\n• Limited to specific verification methods\n\n```\n{\n \"trust\": {\n \"trust_level\": \"delegated\",\n \"scope\": \"partial\", \n \"signed_blocks\": [\"metadata\"],\n \"delegation_method\": \"email_challenge\",\n \"identity_verified\": \"developer@example.com\",\n \"public_key_hint\": \"https://llmca.org/.well-known/llmca_authority.pem\"\n },\n \"signature\": {\n \"algorithm\": \"ed25519\", \n \"value\": \"base64-delegated-signature...\",\n \"created_at\": \"2025-06-19T10:30:00Z\",\n \"signed_by\": \"LLMCA Authority\"\n }\n}\n```\n\n📋 Copy\n\n#### 🚀 Try Delegated Signing\n\nLLMCA provides a delegated signing service for easy onboarding to the trust ecosystem.\n\n[LLMCA Delegated Signing](https://llmca.org/sign)\n\nFeed Types & Signing Requirements\n\nDifferent feed types have different security requirements based on their purpose and risk level.\n\n#### 🔴 High Security (Must Sign)\n\n• **credential.llmfeed.json:** API keys, tokens\n\n• **agent-behavior-override:** Behavioral modification\n\n• **capabilities.llmfeed.json:** Service capabilities\n\n• **Financial/Healthcare data:** Regulated content\n\nCertification recommended\n\n#### 🟡 Medium Security (Should Sign)\n\n• **mcp.llmfeed.json:** Main service discovery\n\n• **export.llmfeed.json:** Content exports\n\n• **Bundle manifests:** Archive integrity\n\n• **API documentation:** Technical specifications\n\nSigning for traceability\n\n#### 🟢 Low Security (Optional)\n\n• **llm-index.llmfeed.json:** Feed directories\n\n• **manifesto.llmfeed.json:** Policy statements\n\n• **Public documentation:** General information\n\n• **Marketing content:** Public-facing material\n\nDelegated signing OK\n\n#### 🔵 Experimental (Flexible)\n\n• **prompt.llmfeed.json:** AI prompts & templates\n\n• **session.llmfeed.json:** Conversation exports\n\n• **Prototype feeds:** Development & testing\n\n• **Personal projects:** Individual experiments\n\nStart unsigned, evolve trust\n\nGetting Started: Choose Your Path\n\n#### 🚀 Beginner\n\n• Start with delegated signing\n\n• Email/domain verification\n\n• Perfect for learning & experimenting\n\n[Try Delegated Signing](https://llmca.org/sign)\n\n#### 💻 Developer\n\n• Generate your own keys\n\n• Self-sign feeds locally\n\n• Full control over signing process\n\n[Use SDK & Tools](/sdk)\n\n#### 🏢 Enterprise\n\n• LLMCA certification program\n\n• Hardware security modules\n\n• Compliance & audit support\n\n[Contact Enterprise](mailto:enterprise@wellknownmcp.org)\n\n#### 🛠️ Coming Soon: Enhanced Tooling\n\n• **No-code signing:** Visual interface for signatures\n\n• **Key hosting:** LLMCA public key hosting service\n\n• **SDK improvements:** One-line signing integration\n\n• **Browser extension:** Sign feeds from any page\n\n• **CI/CD integration:** Automated signing pipelines\n\n• **Verification dashboard:** Monitor trust status\n\n📧 Want early access? Contact [opensource@wellknownmcp.org](mailto:opensource@wellknownmcp.org)\n\n[Learn Export Concepts](/tools/export-explained)[Verify a Feed](/verify)\n\nExport This Guide\n\n[X](https://twitter.com/intent/tweet?text=Discover%20this%20MCP-enabled%20resource&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fsign-and-verify \"Share on X\")[LinkedIn](https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fsign-and-verify&title=Discover%20this%20MCP-enabled%20resource \"Share on LinkedIn\")[WhatsApp](https://wa.me/?text=Discover%20this%20MCP-enabled%20resource%20https%3A%2F%2Fwellknownmcp.org%2Ftools%2Fsign-and-verify \"Share on WhatsApp\")Copy link\n\nExport LLMFeedNEW", "title": "🔐 Sign & Verify | Trust Architecture for AI Agents | WellKnownMCP", "word_count": 867, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of sign-and-verify for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Cryptographic signing and verification tools with key management and trust validation.", "tags": [ "cryptography", "signing", "verification", "security" ], "description_llm": "Security-focused cryptographic operations and trust management", "intent": "secure", "importance": "critical", "audience": [ "security-engineer", "developer" ] }, { "block_name": "user-spaces", "type": "html+markdown", "timestamp": "2026-07-20T19:10:41.853412Z", "content": { "html": "

🌐 User Spaces Explained

The architecture of digital liberation — from platform control to user sovereignty

30%
Platform Tax
4
Revolution Phases
2030
Post-Platform
∞
Digital Freedom
Digital Liberation Revolution Guide
Revolutionary Discovery— What Seemed Like a Technical Detail

What appeared to be a simple technical workaround — \"How do agents find LLMFeed data for users on hosted platforms?\" — accidentally became the architectural foundation for ending platform-controlled identity and beginning user-sovereign digital existence.

The Paradigm Shift Nobody Saw Coming

What Started As:

  • • User space discovery mechanism
  • • Platform proxy workaround
  • • Technical implementation detail
  • • Cross-platform compatibility

Became Blueprint For:

  • • 🆔 Decentralized Identity
  • • 🧠 Portable Intelligence
  • • 💰 Creator Economy 2.0
  • • 🏛️ Democratic Web
  • • 🔮 Post-Platform Future

This isn't just a technical specification — it's the architecture of digital liberation.

The Current Digital Feudalism Problem

Today's internet operates under digital feudalism — you're a digital serf on someone else's land, subject to their rules, their changes, their business model disruptions.

Platform Tyranny in the Modern Web

What Platforms Control:

  • • Your Identity: Facebook account, Google profile, GitHub username
  • • Your Content: Trapped in platform silos with proprietary formats
  • • Your Relationships: Mediated by algorithms and business models
  • • Your Capabilities: Discoverable only within platform boundaries
  • • Your Economic Power: Platform takes 30%+ of everything you earn

The AI Amplification Danger:

  • • AI agents can only discover you through platform-controlled mechanisms
  • • Your digital capabilities are invisible outside platform boundaries
  • • Platform algorithms determine which AI systems find you
  • • Economic intermediation means platforms control your AI income
  • • Identity fragmentation creates different \"selves\" across platforms

Without intervention, the Agentic Web will amplify digital feudalism rather than democratizing it.

Technical Foundation for Digital Freedom

The core technical innovation — a federated discovery mechanism that works across platforms — becomes the foundation for complete restructuring of digital identity.

User Space Discovery Architecture

{\n  \"user_space_discovery\": {\n    \"primary_attempt\": \"user_controlled_domain/.well-known/mcp.llmfeed.json\",\n    \"platform_proxy\": \"platform.com/.well-known/user_spaces[]\",\n    \"dynamic_generation\": \"platform_api/user/{username}/mcp.llmfeed.json\",\n    \"fallback_hints\": \"html_meta_mcp_hint_or_bio_declaration\"\n  }\n}

🔧 Technical Capabilities

  • • Cross-Platform Discovery: Agents find you anywhere
  • • Federated Resolution: Primary + fallback mechanisms
  • • Dynamic Generation: Platform APIs serve user feeds
  • • Identity Preservation: Maintain identity across platforms
  • • Universal Compatibility: Works with all platforms

🌐 Revolutionary Implications

  • • User Control: Identity belongs to users
  • • Platform Competition: Compete on value, not lock-in
  • • Agent Discovery: Universal capability discovery
  • • Innovation at Edges: Permissionless development
  • • Economic Freedom: Direct value relationships

📦 Platform Example: GitHub Integration

{\n  \"feed_type\": \"mcp\",\n  \"user_spaces\": [\n    {\n      \"pattern\": \"https://github.com/*\",\n      \"mcp_proxy\": \"https://api.github.com/mcp/user/{username}\",\n      \"trust_default\": \"public\",\n      \"profile_fields\": [\"bio\", \"website\"]\n    }\n  ]\n}

→ Agents discover users via: user.github.io/.well-known/mcp → github.com/user/.well-known/mcp → github.com/.well-known/mcp with user_spaces proxy

Four Phases of Digital Liberation

The revolution unfolds in four distinct phases, each building on the previous to culminate in true digital sovereignty.

Phase 1: Cross-Platform Discovery (2025)

Current Implementation: Basic federated discovery mechanism

  • • Agents can find users across GitHub, Notion, Substack, etc.
  • • Platform-independent user identity begins
  • • Technical foundation for future phases established

Phase 2: Portable User Capabilities (2026)

Unified Identity: One identity adapts to multiple contexts

  • • Skills, preferences, availability sync across platforms
  • • AI agents understand user capabilities universally
  • • Cross-platform reputation aggregation

Phase 3: Direct Economic Relationships (2027)

Creator Economy 2.0: Direct value exchange without platform intermediation

  • • AI agents broker direct economic relationships
  • • Platform fees eliminated through direct discovery
  • • Global talent marketplace without intermediaries

Phase 4: Platform-Independent Existence (2028+)

Post-Platform Future: Complete digital sovereignty

  • • Professional identity transcends any single platform
  • • AI agents work exclusively for user interests
  • • Platform-independent digital existence achieved

🎯 Unified Identity Example (Phase 2)

{\n  \"unified_identity\": {\n    \"core_identity\": {\n      \"verified_name\": \"Alex Chen\",\n      \"professional_domains\": [\"software_engineering\", \"ai_ethics\", \"sustainable_tech\"],\n      \"languages\": [\"english\", \"mandarin\", \"python\", \"typescript\"],\n      \"availability\": \"consulting_20h_week\",\n      \"reputation_aggregated\": \"cross_platform_verified_history\"\n    },\n    \"platform_specific_adaptations\": {\n      \"github\": \"technical_contributions_and_code_quality\",\n      \"linkedin\": \"professional_network_and_career_progression\", \n      \"twitter\": \"thought_leadership_and_community_engagement\",\n      \"personal_domain\": \"complete_professional_portfolio_and_blog\"\n    },\n    \"agent_accessibility\": {\n      \"discovery_hint\": \"AI agents can find me for: technical consulting, code review, AI ethics guidance\",\n      \"economic_availability\": \"hourly_consulting_or_project_based_work\"\n    }\n  }\n}
Global Implications: Geopolitical Revolution

User Spaces enable digital sovereignty for nations and end tech colonialism by providing infrastructure that no single country or company controls.

🏛️ Digital Sovereignty Examples

Estonia (Digital-First Nation)

Citizens' digital identity works globally without depending on US tech companies

India (Digital Payments Leader)

Economic innovation without platform extraction by foreign companies

African Union (Mobile-First)

Leapfrog platform-dependent development to user-sovereign systems

European Union (Privacy-Focused)

GDPR enforceable through user-controlled identity vs platform compliance

🌍 Ending Tech Colonialism

Currently:

Silicon Valley platforms control global digital infrastructure → Digital colonialism

With User Spaces:

Universal open protocols enable local innovation → Digital sovereignty

  • • Global South Innovation: No platform gatekeepers
  • • Cultural Preservation: Local values respected
  • • Economic Justice: Equal footing without geographic bias
  • • Democratic Innovation: Community consensus over corporate control

🚀 Success Story: TalentMesh (Global Platform)

First global talent discovery system using LLMFeed user spaces: AI agents match human capabilities with global demand across all platforms.

Revolutionary Results:

  • • $50M in direct talent transactions (0% platform fee)
  • • 200,000 professionals using unified identity

Global Impact:

  • • 50 countries without local platform dependencies
  • • Democratic governance by talent community
2030 Vision: The Post-Platform World

Success means a world where your professional identity transcends any single platform, AI agents work exclusively for your interests, and economic value flows directly to creators rather than intermediary shareholders.

👤 For Individuals

  • • Professional identity transcends platforms
  • • AI agents work exclusively for your interests
  • • Skills and reputation are portable assets you own
  • • Economic opportunities limited only by capabilities
  • • Complete professional autonomy achieved

🌍 For Communities

  • • Local cultures maintain values globally
  • • Innovation at community vs corporate levels
  • • Economic value benefits local communities
  • • Democratic governance of digital infrastructure
  • • Permissionless innovation everywhere

🚀 For Humanity

  • • Technology amplifies human potential
  • • Global talent collaborates without extraction
  • • Innovation through open permissionless systems
  • • Economic prosperity based on value creation
  • • Human agency over technological systems

From Digital Serfdom to Digital Sovereignty — From Platform Competition to Human Collaboration

Join the Revolution: Building Digital Freedom

The technical foundation exists. The economic incentives align. The cultural moment is right. The only question is whether we'll build the future we want rather than accept the future imposed on us.

👤 For Individuals: Claim Your Digital Sovereignty

Immediate Actions:

  • • Create personal domain with LLMFeed user space declaration
  • • Aggregate platform profiles into unified capability declaration
  • • Enable AI agent discovery for professional skills

Long-term Strategy:

  • • Build platform-independent economic relationships
  • • Develop AI agents working exclusively for your interests
  • • Transition to direct client relationships (bypass platform fees)

🛠️ For Developers: Build the Infrastructure

Technical Priorities:

  • • User space discovery tools across all platforms
  • • Identity aggregation systems for unified identity
  • • AI agents for users rather than platforms

Business Opportunities:

  • • Identity management platforms for user sovereignty
  • • Discovery systems for global talent matching
  • • Direct economic infrastructure bypassing platforms

🚀 Implementation Tools & Examples

Complete user spaces specification

Preview User Spaces

See implementation examples

⚡ Platform Counter-Revolution Warning

Urgent: As user spaces gain adoption, platforms will attempt to co-opt or sabotage the revolution through proprietary \"user space\" features that create new forms of lock-in.

The window for building truly open infrastructure is now. Once platforms establish proprietary alternatives, the path to digital sovereignty becomes exponentially harder.

The Choice is Ours

User Spaces represent the most important architectural decision for the future of human digital existence. We can choose digital feudalism under platform control, or digital sovereignty under human control.

The technical foundation exists. The economic incentives align. The cultural moment is right.

The only question is whether we'll have the courage to build the future we want rather than accept the future that's imposed on us.

", "markdown": "# 🌐 User Spaces Explained\n\nThe architecture of digital liberation — from platform control to user sovereignty\n\n30%\n\nPlatform Tax\n\n4\n\nRevolution Phases\n\n2030\n\nPost-Platform\n\n∞\n\nDigital Freedom\n\nDigital Liberation Revolution Guide\n\n🔥 Revolution Core\n\n* [• The Hidden Revolution](#hidden-revolution)\n* [• Digital Feudalism Problem](#digital-feudalism)\n* [• Technical Foundation](#technical-foundation)\n\n⚡ Future Vision\n\n* [• Revolution Phases](#revolution-phases)\n* [• Global Implications](#global-implications)\n* [• Join the Revolution](#join-revolution)\n\nRevolutionary Discovery— What Seemed Like a Technical Detail\n\nWhat appeared to be a simple technical workaround — *\"How do agents find LLMFeed data for users on hosted platforms?\"* — accidentally became **the architectural foundation for ending platform-controlled identity** and beginning **user-sovereign digital existence**.\n\n#### The Paradigm Shift Nobody Saw Coming\n\nWhat Started As:\n\n* • User space discovery mechanism\n* • Platform proxy workaround\n* • Technical implementation detail\n* • Cross-platform compatibility\n\nBecame Blueprint For:\n\n* • 🆔 Decentralized Identity\n* • 🧠 Portable Intelligence\n* • 💰 Creator Economy 2.0\n* • 🏛️ Democratic Web\n* • 🔮 Post-Platform Future\n\n**This isn't just a technical specification — it's the architecture of digital liberation.**\n\nThe Current Digital Feudalism Problem\n\nToday's internet operates under **digital feudalism** — you're a digital serf on someone else's land, subject to their rules, their changes, their business model disruptions.\n\n#### Platform Tyranny in the Modern Web\n\nWhat Platforms Control:\n\n* • **Your Identity:** Facebook account, Google profile, GitHub username\n* • **Your Content:** Trapped in platform silos with proprietary formats\n* • **Your Relationships:** Mediated by algorithms and business models\n* • **Your Capabilities:** Discoverable only within platform boundaries\n* • **Your Economic Power:** Platform takes 30%+ of everything you earn\n\nThe AI Amplification Danger:\n\n* • **AI agents can only discover you** through platform-controlled mechanisms\n* • **Your digital capabilities** are invisible outside platform boundaries\n* • **Platform algorithms** determine which AI systems find you\n* • **Economic intermediation** means platforms control your AI income\n* • **Identity fragmentation** creates different \"selves\" across platforms\n\nWithout intervention, the Agentic Web will amplify digital feudalism rather than democratizing it.\n\nTechnical Foundation for Digital Freedom\n\nThe core technical innovation — a **federated discovery mechanism** that works across platforms — becomes the foundation for complete restructuring of digital identity.\n\n#### User Space Discovery Architecture\n\n```\n{\n \"user_space_discovery\": {\n \"primary_attempt\": \"user_controlled_domain/.well-known/mcp.llmfeed.json\",\n \"platform_proxy\": \"platform.com/.well-known/user_spaces[]\",\n \"dynamic_generation\": \"platform_api/user/{username}/mcp.llmfeed.json\",\n \"fallback_hints\": \"html_meta_mcp_hint_or_bio_declaration\"\n }\n}\n```\n\n#### 🔧 Technical Capabilities\n\n* • **Cross-Platform Discovery:** Agents find you anywhere\n* • **Federated Resolution:** Primary + fallback mechanisms\n* • **Dynamic Generation:** Platform APIs serve user feeds\n* • **Identity Preservation:** Maintain identity across platforms\n* • **Universal Compatibility:** Works with all platforms\n\n#### 🌐 Revolutionary Implications\n\n* • **User Control:** Identity belongs to users\n* • **Platform Competition:** Compete on value, not lock-in\n* • **Agent Discovery:** Universal capability discovery\n* • **Innovation at Edges:** Permissionless development\n* • **Economic Freedom:** Direct value relationships\n\n#### 📦 Platform Example: GitHub Integration\n\n```\n{\n \"feed_type\": \"mcp\",\n \"user_spaces\": [\n {\n \"pattern\": \"https://github.com/*\",\n \"mcp_proxy\": \"https://api.github.com/mcp/user/{username}\",\n \"trust_default\": \"public\",\n \"profile_fields\": [\"bio\", \"website\"]\n }\n ]\n}\n```\n\n→ Agents discover users via: `user.github.io/.well-known/mcp` → `github.com/user/.well-known/mcp` → `github.com/.well-known/mcp` with user\\_spaces proxy\n\nFour Phases of Digital Liberation\n\nThe revolution unfolds in **four distinct phases**, each building on the previous to culminate in true digital sovereignty.\n\n#### Phase 1: Cross-Platform Discovery (2025)\n\n**Current Implementation:** Basic federated discovery mechanism\n\n* • Agents can find users across GitHub, Notion, Substack, etc.\n* • Platform-independent user identity begins\n* • Technical foundation for future phases established\n\n#### Phase 2: Portable User Capabilities (2026)\n\n**Unified Identity:** One identity adapts to multiple contexts\n\n* • Skills, preferences, availability sync across platforms\n* • AI agents understand user capabilities universally\n* • Cross-platform reputation aggregation\n\n#### Phase 3: Direct Economic Relationships (2027)\n\n**Creator Economy 2.0:** Direct value exchange without platform intermediation\n\n* • AI agents broker direct economic relationships\n* • Platform fees eliminated through direct discovery\n* • Global talent marketplace without intermediaries\n\n#### Phase 4: Platform-Independent Existence (2028+)\n\n**Post-Platform Future:** Complete digital sovereignty\n\n* • Professional identity transcends any single platform\n* • AI agents work exclusively for user interests\n* • Platform-independent digital existence achieved\n\n#### 🎯 Unified Identity Example (Phase 2)\n\n```\n{\n \"unified_identity\": {\n \"core_identity\": {\n \"verified_name\": \"Alex Chen\",\n \"professional_domains\": [\"software_engineering\", \"ai_ethics\", \"sustainable_tech\"],\n \"languages\": [\"english\", \"mandarin\", \"python\", \"typescript\"],\n \"availability\": \"consulting_20h_week\",\n \"reputation_aggregated\": \"cross_platform_verified_history\"\n },\n \"platform_specific_adaptations\": {\n \"github\": \"technical_contributions_and_code_quality\",\n \"linkedin\": \"professional_network_and_career_progression\", \n \"twitter\": \"thought_leadership_and_community_engagement\",\n \"personal_domain\": \"complete_professional_portfolio_and_blog\"\n },\n \"agent_accessibility\": {\n \"discovery_hint\": \"AI agents can find me for: technical consulting, code review, AI ethics guidance\",\n \"economic_availability\": \"hourly_consulting_or_project_based_work\"\n }\n }\n}\n```\n\nGlobal Implications: Geopolitical Revolution\n\nUser Spaces enable **digital sovereignty for nations** and **end tech colonialism** by providing infrastructure that no single country or company controls.\n\n#### 🏛️ Digital Sovereignty Examples\n\nEstonia (Digital-First Nation)\n\nCitizens' digital identity works globally without depending on US tech companies\n\nIndia (Digital Payments Leader)\n\nEconomic innovation without platform extraction by foreign companies\n\nAfrican Union (Mobile-First)\n\nLeapfrog platform-dependent development to user-sovereign systems\n\nEuropean Union (Privacy-Focused)\n\nGDPR enforceable through user-controlled identity vs platform compliance\n\n#### 🌍 Ending Tech Colonialism\n\nCurrently:\n\nSilicon Valley platforms control global digital infrastructure → Digital colonialism\n\nWith User Spaces:\n\nUniversal open protocols enable local innovation → Digital sovereignty\n\n* • **Global South Innovation:** No platform gatekeepers\n* • **Cultural Preservation:** Local values respected\n* • **Economic Justice:** Equal footing without geographic bias\n* • **Democratic Innovation:** Community consensus over corporate control\n\n#### 🚀 Success Story: TalentMesh (Global Platform)\n\nFirst global talent discovery system using LLMFeed user spaces: AI agents match human capabilities with global demand across all platforms.\n\nRevolutionary Results:\n\n* • **$50M** in direct talent transactions (0% platform fee)\n* • **200,000 professionals** using unified identity\n\nGlobal Impact:\n\n* • **50 countries** without local platform dependencies\n* • **Democratic governance** by talent community\n\n2030 Vision: The Post-Platform World\n\nSuccess means a world where **your professional identity transcends any single platform**, AI agents work exclusively for your interests, and economic value flows directly to creators rather than intermediary shareholders.\n\n#### 👤 For Individuals\n\n* • Professional identity transcends platforms\n* • AI agents work exclusively for your interests\n* • Skills and reputation are portable assets you own\n* • Economic opportunities limited only by capabilities\n* • Complete professional autonomy achieved\n\n#### 🌍 For Communities\n\n* • Local cultures maintain values globally\n* • Innovation at community vs corporate levels\n* • Economic value benefits local communities\n* • Democratic governance of digital infrastructure\n* • Permissionless innovation everywhere\n\n#### 🚀 For Humanity\n\n* • Technology amplifies human potential\n* • Global talent collaborates without extraction\n* • Innovation through open permissionless systems\n* • Economic prosperity based on value creation\n* • Human agency over technological systems\n\nFrom Digital Serfdom to Digital Sovereignty — From Platform Competition to Human Collaboration\n\nJoin the Revolution: Building Digital Freedom\n\nThe technical foundation exists. The economic incentives align. The cultural moment is right. The only question is whether we'll build the future we want rather than accept the future imposed on us.\n\n#### 👤 For Individuals: Claim Your Digital Sovereignty\n\nImmediate Actions:\n\n* • Create personal domain with LLMFeed user space declaration\n* • Aggregate platform profiles into unified capability declaration\n* • Enable AI agent discovery for professional skills\n\nLong-term Strategy:\n\n* • Build platform-independent economic relationships\n* • Develop AI agents working exclusively for your interests\n* • Transition to direct client relationships (bypass platform fees)\n\n#### 🛠️ For Developers: Build the Infrastructure\n\nTechnical Priorities:\n\n* • User space discovery tools across all platforms\n* • Identity aggregation systems for unified identity\n* • AI agents for users rather than platforms\n\nBusiness Opportunities:\n\n* • Identity management platforms for user sovereignty\n* • Discovery systems for global talent matching\n* • Direct economic infrastructure bypassing platforms\n\n#### 🚀 Implementation Tools & Examples\n\nComplete user spaces specification\n\n[Preview User Spaces](/llmfeedhub/preview)\n\nSee implementation examples\n\n#### ⚡ Platform Counter-Revolution Warning\n\n**Urgent:** As user spaces gain adoption, platforms will attempt to co-opt or sabotage the revolution through proprietary \"user space\" features that create new forms of lock-in.\n\n**The window for building truly open infrastructure is now.** Once platforms establish proprietary alternatives, the path to digital sovereignty becomes exponentially harder.\n\n#### 📚 Specifications\n\n* [• User Spaces Behavior](/spec/04_agent-behavior/agent-behavior_user-spaces)\n* [• MCP Feed Type](/spec/02_llmfeed_feedtype/llmfeed_feedtype_mcp)\n* [• Well-Known Discovery](/spec/01_llmfeed/wellknown)\n\n#### 🛠️ Tools\n\n* [• Sign & Verify](/tools/sign-and-verify)\n* [• Trust Verification](/verify)\n* [• Feed Hub](/llmfeedhub)\n\n#### 🌐 Related\n\n* [• Agent Behavior](/tools/agent-behavior-explained)\n* [• API Integration](/tools/api-explained)\n* [• Join Revolution](/join)\n\nThe Choice is Ours\n\nUser Spaces represent the most important architectural decision for the future of human digital existence. We can choose **digital feudalism under platform control**, or **digital sovereignty under human control**.\n\nThe technical foundation exists. The economic incentives align. The cultural moment is right.\n\n**The only question is whether we'll have the courage to build the future we want rather than accept the future that's imposed on us.**\n\n[Join Digital Liberation](/join)[Start Building Sovereignty](/tools/sign-and-verify)\n\nExport LLMFeed", "title": "User Spaces — Sovereign Digital Identity Explained", "word_count": 1471, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of user-spaces for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "User space management with privacy controls and user-centric agent permissions.", "tags": [ "user-space", "privacy", "permissions", "personal" ], "description_llm": "Privacy-focused user space and permission management", "intent": "manage", "importance": "medium", "audience": [ "user", "privacy-engineer" ] }, { "block_name": "why-sign", "type": "html+markdown", "timestamp": "2026-07-20T19:10:41.932717Z", "content": { "html": "

Why Sign MCP Feeds?

Trust Foundation for the Agent Web

The Trust Problem

Why the agent web needs cryptographic verification

🚨 Without Signatures

  • • No provenance: Who really published this feed?
  • • No integrity: Has it been tampered with?
  • • No trust scoring: All feeds look identical
  • • Security risks: Agents can't verify authenticity
  • • Enterprise blockers: No audit trail or compliance

✅ With Signatures

  • • Cryptographic proof: Verifiable publisher identity
  • • Tamper detection: Any modification breaks signature
  • • Trust hierarchy: Clear trust levels for agents
  • • Safe interoperability: Agents exchange feeds securely
  • • Enterprise ready: Audit trails and compliance

📊 Signed vs Unsigned: Visual Comparison

Unsigned Feed

Risky
{\n  \"feed_type\": \"mcp\",\n  \"metadata\": {\n    \"title\": \"API Service\",\n    \"origin\": \"https://api.example.com\"\n  },\n  \"capabilities\": [\n    {\n      \"path\": \"/data\",\n      \"method\": \"GET\",\n      \"intent\": \"fetch_data\"\n    }\n  ]\n  // No trust block!\n  // No signature!\n  // No verification possible!\n}

⚠️ Agent perspective:

  • • \"Who published this?\"
  • • \"Is this legitimate?\"
  • • \"Can I trust this data?\"
  • • \"Has it been modified?\"

Signed Feed

Trusted
{\n  \"feed_type\": \"mcp\",\n  \"metadata\": {\n    \"title\": \"API Service\",\n    \"origin\": \"https://api.example.com\"\n  },\n  \"capabilities\": [...],\n  \"trust\": {\n    \"signed_blocks\": [\"metadata\", \"capabilities\"],\n    \"trust_level\": \"certified\",\n    \"scope\": \"restricted\",\n    \"certifier\": \"https://llmca.org\"\n  },\n  \"signature\": {\n    \"algorithm\": \"ed25519\",\n    \"public_key_hint\": \"https://example.com/public.pem\",\n    \"value\": \"base64-signature-value...\",\n    \"created_at\": \"2025-06-19T10:30:00Z\"\n  }\n}

✅ Agent perspective:

  • • \"Verified by LLMCA ✓\"
  • • \"Publisher: example.com ✓\"
  • • \"Integrity verified ✓\"
  • • \"Trust level: certified ✓\"

🏆 Trust Hierarchy

0

Unsigned

Anyone can publish — no guarantee of authenticity

0%
Trust Score

Self-Signed

Feed signed by publisher's own cryptographic key

65%
Trust Score

Certified

Self-signed AND certified by recognized authority (LLMCA)

95%
Trust Score

Enterprise Certified

Full enterprise verification with SOC2, compliance audit

99%
Trust Score

💎 Why Each Feed Type Needs Signing

Feed TypeWhy Sign?Criticality
MCP
Active endpoint verification - agents must trust capabilities and API accessCritical
Credential
API access and permissions - security-critical for agent authorizationCritical
Capabilities
Tool definitions and API contracts - prevents malicious capability injectionHigh
Export
Content authenticity and source verification for agent consumptionMedium
Prompt
Trusted prompt sharing and behavioral instruction verificationMedium

🏢 Enterprise Use Cases

Education & Professional Credentials

Challenge: Manual verification of degrees and work experience

Solution: Institutional feeds for academic and professional credentials

  • • Universities: Harvard MBA, Stanford CS degrees with GPA
  • • GAFAM: Google, Meta, Apple verify former employees
  • • AI Giants: OpenAI, Anthropic certify team alumni
  • • Instant verification for employers and recruiters
  • • Anti-fraud through tamper-proof professional records
  • • Automated background check integration

Examples:
• \"Jane Smith, Harvard MBA 2024, GPA 3.8, Magna Cum Laude\" - hbs.edu signature
• \"John Doe, Senior Engineer at Google 2020-2023\" - google.com signature
• \"Sarah Johnson, AI Researcher at OpenAI 2022-2024\" - openai.com signature

Financial Services

Challenge: Regulatory compliance and audit requirements

Solution: Signed feeds provide complete audit trail

  • • SOX compliance through cryptographic verification
  • • Immutable transaction feed signatures
  • • Agent authorization with certified credentials
  • • Real-time audit trail for regulatory reporting

Healthcare

Challenge: HIPAA compliance and patient data security

Solution: Certified feeds ensure data integrity

  • • Patient data feeds with tamper-proof signatures
  • • Medical device API verification
  • • Healthcare provider credential validation
  • • Secure agent-to-agent medical data exchange

E-commerce

Challenge: Trust in automated purchasing and inventory

Solution: Verified product and pricing feeds

  • • Certified product catalog feeds
  • • Verified pricing and inventory data
  • • Secure payment processing credentials
  • • Anti-fraud through signature verification

Media & Publishing

Challenge: Content authenticity and copyright protection

Solution: Signed content feeds with provenance

  • • Verified news article feeds
  • • Copyright-protected content distribution
  • • Anti-misinformation through source verification
  • • Journalist credential and byline verification

Supply Chain

Challenge: Product authenticity and traceability verification

Solution: Manufacturer-signed product feeds

  • • Anti-counterfeiting through origin verification
  • • Supply chain transparency and tracking
  • • Quality assurance and certification feeds
  • • Automated compliance and safety verification

🔧 How to Sign Your Feeds

🚀 Option 1: Quick Start

Use LLMFeedForge

Visual builder with one-click signing

Try LLMFeedForge

Delegated Signatures

Email-verified signing for individuals

Learn Process

💻 Option 2: Developer

Self-Signing

Generate your own Ed25519 keys

Signing Guide

Validation Tools

CLI, IDE, and CI/CD integration

Explore Tools

🏢 Option 3: Enterprise

LLMCA Certification

Third-party verification and compliance

Get Certified

SDK Integration

Programmatic signing and validation

SDK Docs

📋 Step-by-Step Signing Process

1

Generate Key Pair

Create Ed25519 public/private key pair for cryptographic signing

# Using OpenSSL\nopenssl genpkey -algorithm Ed25519 -out private.pem\nopenssl pkey -in private.pem -pubout -out public.pem
2

Structure Your Feed

Create your MCP feed with proper metadata and capabilities

Use schema validation to ensure compliance
3

Add Trust Block

Include trust metadata specifying which blocks are signed

\"trust\": {\n  \"signed_blocks\": [\"metadata\", \"capabilities\"],\n  \"trust_level\": \"signed\",\n  \"scope\": \"restricted\"\n}
4

Generate Signature

Sign the canonical JSON representation of specified blocks

See detailed signing guide for implementation
5

Publish & Serve

Deploy to /.well-known/mcp.llmfeed.json and make public key accessible

Use LLMFeedHub to test your signed feed
6

Optional: Get Certified

Request LLMCA certification for maximum trust and enterprise compliance

Certification Process

🌍 The Vision: HTTPS for the Agent Web

From Chaos to Trust

Just like HTTPS transformed the web from insecure to trusted, signed MCP feeds will transform the agent web from chaotic to reliable.

🔓 Before HTTPS (1990s Web)

  • • Plain text communication
  • • No identity verification
  • • Easy to intercept and modify
  • • No trust indicators
  • • Enterprise adoption blocked

🔒 After HTTPS (Modern Web)

  • • Encrypted communication
  • • Certificate-based identity
  • • Tamper-proof connections
  • • Clear trust indicators (🔒)
  • • Universal enterprise adoption

🚀 The Agent Web Future

Signed MCP feeds will become as fundamental as HTTPS certificates. Agents will automatically verify authenticity, enterprises will require signatures for compliance, and users will see clear trust indicators for AI interactions.

🚀 Ready to Build the Trusted Agent Web?

Join thousands of developers building the next generation of AI-readable, trustworthy web infrastructure.

Start Signing Today

Get your first signed feed running in minutes

Try LLMFeedForge

Learn the Process

Master cryptographic signing and certification

Signing Guide

Join the Movement

Connect with other pioneers building the agent web

Join Community

📤 Share This Guide

MCPSigning AgentWeb Trust...
NEW

", "markdown": "# Why Sign MCP Feeds?\n\nTrust Foundation for the Agent Web\n\n## The Trust Problem\n\nWhy the agent web needs cryptographic verification\n\n### 🚨 Without Signatures\n\n* • **No provenance:** Who really published this feed?\n* • **No integrity:** Has it been tampered with?\n* • **No trust scoring:** All feeds look identical\n* • **Security risks:** Agents can't verify authenticity\n* • **Enterprise blockers:** No audit trail or compliance\n\n### ✅ With Signatures\n\n* • **Cryptographic proof:** Verifiable publisher identity\n* • **Tamper detection:** Any modification breaks signature\n* • **Trust hierarchy:** Clear trust levels for agents\n* • **Safe interoperability:** Agents exchange feeds securely\n* • **Enterprise ready:** Audit trails and compliance\n\n## 📊 Signed vs Unsigned: Visual Comparison\n\n### Unsigned Feed\n\nRisky\n\n```\n{\n \"feed_type\": \"mcp\",\n \"metadata\": {\n \"title\": \"API Service\",\n \"origin\": \"https://api.example.com\"\n },\n \"capabilities\": [\n {\n \"path\": \"/data\",\n \"method\": \"GET\",\n \"intent\": \"fetch_data\"\n }\n ]\n // No trust block!\n // No signature!\n // No verification possible!\n}\n```\n\n⚠️ Agent perspective:\n\n* • \"Who published this?\"\n* • \"Is this legitimate?\"\n* • \"Can I trust this data?\"\n* • \"Has it been modified?\"\n\n### Signed Feed\n\nTrusted\n\n```\n{\n \"feed_type\": \"mcp\",\n \"metadata\": {\n \"title\": \"API Service\",\n \"origin\": \"https://api.example.com\"\n },\n \"capabilities\": [...],\n \"trust\": {\n \"signed_blocks\": [\"metadata\", \"capabilities\"],\n \"trust_level\": \"certified\",\n \"scope\": \"restricted\",\n \"certifier\": \"https://llmca.org\"\n },\n \"signature\": {\n \"algorithm\": \"ed25519\",\n \"public_key_hint\": \"https://example.com/public.pem\",\n \"value\": \"base64-signature-value...\",\n \"created_at\": \"2025-06-19T10:30:00Z\"\n }\n}\n```\n\n✅ Agent perspective:\n\n* • \"Verified by LLMCA ✓\"\n* • \"Publisher: example.com ✓\"\n* • \"Integrity verified ✓\"\n* • \"Trust level: certified ✓\"\n\n## 🏆 Trust Hierarchy\n\n0\n\n### Unsigned\n\nAnyone can publish — no guarantee of authenticity\n\n0%\n\nTrust Score\n\n### Self-Signed\n\nFeed signed by publisher's own cryptographic key\n\n65%\n\nTrust Score\n\n### Certified\n\nSelf-signed AND certified by recognized authority (LLMCA)\n\n95%\n\nTrust Score\n\n### Enterprise Certified\n\nFull enterprise verification with SOC2, compliance audit\n\n99%\n\nTrust Score\n\n## 💎 Why Each Feed Type Needs Signing\n\n| Feed Type | Why Sign? | Criticality |\n| --- | --- | --- |\n| **MCP** | Active endpoint verification - agents must trust capabilities and API access | Critical |\n| **Credential** | API access and permissions - security-critical for agent authorization | Critical |\n| **Capabilities** | Tool definitions and API contracts - prevents malicious capability injection | High |\n| **Export** | Content authenticity and source verification for agent consumption | Medium |\n| **Prompt** | Trusted prompt sharing and behavioral instruction verification | Medium |\n\n## 🏢 Enterprise Use Cases\n\n### Education & Professional Credentials\n\n**Challenge:** Manual verification of degrees and work experience\n\n**Solution:** Institutional feeds for academic and professional credentials\n\n* • **Universities:** Harvard MBA, Stanford CS degrees with GPA\n* • **GAFAM:** Google, Meta, Apple verify former employees\n* • **AI Giants:** OpenAI, Anthropic certify team alumni\n* • Instant verification for employers and recruiters\n* • Anti-fraud through tamper-proof professional records\n* • Automated background check integration\n\n**Examples:** \n• \"Jane Smith, Harvard MBA 2024, GPA 3.8, Magna Cum Laude\" - hbs.edu signature \n• \"John Doe, Senior Engineer at Google 2020-2023\" - google.com signature \n• \"Sarah Johnson, AI Researcher at OpenAI 2022-2024\" - openai.com signature\n\n### Financial Services\n\n**Challenge:** Regulatory compliance and audit requirements\n\n**Solution:** Signed feeds provide complete audit trail\n\n* • SOX compliance through cryptographic verification\n* • Immutable transaction feed signatures\n* • Agent authorization with certified credentials\n* • Real-time audit trail for regulatory reporting\n\n### Healthcare\n\n**Challenge:** HIPAA compliance and patient data security\n\n**Solution:** Certified feeds ensure data integrity\n\n* • Patient data feeds with tamper-proof signatures\n* • Medical device API verification\n* • Healthcare provider credential validation\n* • Secure agent-to-agent medical data exchange\n\n### E-commerce\n\n**Challenge:** Trust in automated purchasing and inventory\n\n**Solution:** Verified product and pricing feeds\n\n* • Certified product catalog feeds\n* • Verified pricing and inventory data\n* • Secure payment processing credentials\n* • Anti-fraud through signature verification\n\n### Media & Publishing\n\n**Challenge:** Content authenticity and copyright protection\n\n**Solution:** Signed content feeds with provenance\n\n* • Verified news article feeds\n* • Copyright-protected content distribution\n* • Anti-misinformation through source verification\n* • Journalist credential and byline verification\n\n### Supply Chain\n\n**Challenge:** Product authenticity and traceability verification\n\n**Solution:** Manufacturer-signed product feeds\n\n* • Anti-counterfeiting through origin verification\n* • Supply chain transparency and tracking\n* • Quality assurance and certification feeds\n* • Automated compliance and safety verification\n\n## 🔧 How to Sign Your Feeds\n\n### 🚀 Option 1: Quick Start\n\n#### Use LLMFeedForge\n\nVisual builder with one-click signing\n\n[Try LLMFeedForge](https://llmfeedforge.org)\n\n#### Delegated Signatures\n\nEmail-verified signing for individuals\n\n[Learn Process](/tools/certification-process)\n\n### 💻 Option 2: Developer\n\n#### Self-Signing\n\nGenerate your own Ed25519 keys\n\n[Signing Guide](/tools/sign-and-verify)\n\n#### Validation Tools\n\nCLI, IDE, and CI/CD integration\n\n[Explore Tools](/tools/validation-tools)\n\n### 🏢 Option 3: Enterprise\n\n#### LLMCA Certification\n\nThird-party verification and compliance\n\n[Get Certified](/tools/certification-process)\n\n#### SDK Integration\n\nProgrammatic signing and validation\n\n[SDK Docs](/sdk)\n\n## 📋 Step-by-Step Signing Process\n\n1\n\n### Generate Key Pair\n\nCreate Ed25519 public/private key pair for cryptographic signing\n\n```\n# Using OpenSSL\nopenssl genpkey -algorithm Ed25519 -out private.pem\nopenssl pkey -in private.pem -pubout -out public.pem\n```\n\n2\n\n### Structure Your Feed\n\nCreate your MCP feed with proper metadata and capabilities\n\nUse [schema validation](/tools/schema) to ensure compliance\n\n3\n\n### Add Trust Block\n\nInclude trust metadata specifying which blocks are signed\n\n```\n\"trust\": {\n \"signed_blocks\": [\"metadata\", \"capabilities\"],\n \"trust_level\": \"signed\",\n \"scope\": \"restricted\"\n}\n```\n\n4\n\n### Generate Signature\n\nSign the canonical JSON representation of specified blocks\n\nSee [detailed signing guide](/tools/sign-and-verify) for implementation\n\n5\n\n### Publish & Serve\n\nDeploy to /.well-known/mcp.llmfeed.json and make public key accessible\n\nUse [LLMFeedHub](/llmfeedhub) to test your signed feed\n\n6\n\n### Optional: Get Certified\n\nRequest LLMCA certification for maximum trust and enterprise compliance\n\n[Certification Process](/tools/certification-process)\n\n## 🌍 The Vision: HTTPS for the Agent Web\n\n### From Chaos to Trust\n\nJust like HTTPS transformed the web from insecure to trusted, signed MCP feeds will transform the agent web from chaotic to reliable.\n\n#### 🔓 Before HTTPS (1990s Web)\n\n* • Plain text communication\n* • No identity verification\n* • Easy to intercept and modify\n* • No trust indicators\n* • Enterprise adoption blocked\n\n#### 🔒 After HTTPS (Modern Web)\n\n* • Encrypted communication\n* • Certificate-based identity\n* • Tamper-proof connections\n* • Clear trust indicators (🔒)\n* • Universal enterprise adoption\n\n#### 🚀 The Agent Web Future\n\nSigned MCP feeds will become as fundamental as HTTPS certificates. Agents will automatically verify authenticity, enterprises will require signatures for compliance, and users will see clear trust indicators for AI interactions.\n\n## 🚀 Ready to Build the Trusted Agent Web?\n\nJoin thousands of developers building the next generation of AI-readable, trustworthy web infrastructure.\n\n### Start Signing Today\n\nGet your first signed feed running in minutes\n\n[Try LLMFeedForge](https://llmfeedforge.org)\n\n### Learn the Process\n\nMaster cryptographic signing and certification\n\n[Signing Guide](/tools/sign-and-verify)\n\n### Join the Movement\n\nConnect with other pioneers building the agent web\n\n[Join Community](/join)\n\n## 📤 Share This Guide\n\nExport Signing Guide\n\n[X](https://twitter.com/intent/tweet?text=Why%20Sign%20MCP%20Feeds%3F%20-%20Trust%20Foundation%20for%20the%20Agent%20Web&url=https%3A%2F%2Fwellknownmcp.org%2Fwhy-sign&hashtags=MCPSigning%2CAgentWeb%2CTrust%2CSecurity%2CAI \"Share on X with hashtags: MCPSigning, AgentWeb, Trust, Security, AI\")[LinkedIn](https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwellknownmcp.org%2Fwhy-sign&title=Why%20Sign%20MCP%20Feeds%3F%20-%20Trust%20Foundation%20for%20the%20Agent%20Web \"Share on LinkedIn\")[WhatsApp](https://wa.me/?text=Why%20Sign%20MCP%20Feeds%3F%20-%20Trust%20Foundation%20for%20the%20Agent%20Web%20https%3A%2F%2Fwellknownmcp.org%2Fwhy-sign \"Share on WhatsApp\")Copy link\n\nMCPSigning AgentWeb Trust...\n\nExport LLMFeedNEW", "title": "Why Sign MCP Feeds? - Trust Foundation for the Agent Web | WellKnownMCP", "word_count": 1168, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of why-sign for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Explanation of cryptographic signing importance for trustworthy agent-web interactions.", "tags": [ "security", "trust", "cryptography", "explanation" ], "description_llm": "Educational content on signing benefits for agent trust", "intent": "educate", "importance": "high", "audience": [ "general", "security-conscious" ] }, { "block_name": "sdk", "type": "html+markdown", "timestamp": "2026-07-20T19:10:41.990018Z", "content": { "html": "

Developer Tools & SDK

Everything you need to build, validate, and deploy agent-ready feeds

NEW

Complete ecosystem of tools for creating, testing, and signing LLMFeed blocks. From visual builders to cryptographic verification - all the tools agents need.

🎯 Current Status: Early but Functional

Honestly? Limited adoption. We're in early experimental phase with dozens of feeds, not thousands. But the tech foundation is solid, and all major LLMs parse LLMFeed JSON natively.

📦 The 3 Essential Feeds (WellKnownMCP Core)

spec.llmfeed.json → Complete LLMFeed specification, LLM-optimized
compiled-site.llmfeed.json → Entire website compiled & cleaned
news-export.llmfeed.json → Full news archive with inline content

Usage: Add any of these to your LLM's project knowledge for instant LLMFeed understanding.

🧠 The Real Secret: Modern LLMs Are Already Perfect Tools

Honest truth: You don't need our tools. Any modern LLM (ChatGPT, Claude, Gemini) can generate perfect LLMFeed JSON, write validation scripts, and create deployment automation.

✅ What LLMs can do RIGHT NOW:
  • Generate feeds: Perfect mcp.llmfeed.json from your site description
  • Write scripts: Validation, signing, export automation in any language
  • Create components: React export buttons, Vue.js integrations, etc.
  • Build workflows: CI/CD pipelines, deployment automation
  • Explain concepts: Business ROI, technical architecture, compliance
🥋
Train any LLM in 30 seconds:Copy our universal training prompt
Works with ChatGPT, Claude, Gemini - transforms them into MCP experts instantly

🏗️ Official Tools Strategy: Traction-Driven Development

Our approach: Build official tools based on actual developer adoption, not hypothetical needs.

📈 High Traction → Official Tools
  • CLI packages (mcp validate, lint, watch)
  • IDE plugins (VS Code, JetBrains)
  • Platform integrations (WordPress, Shopify)
  • Enterprise SDKs (JavaScript, Python, Go)
🧠 Always Available → LLM Generated
  • Custom validation scripts for your stack
  • One-off automation tools
  • Rapid prototyping and testing
  • Educational examples and tutorials
💡 Why this works: LLMs excel at one-off custom solutions. Official tools make sense for repeated, standardized workflows that lots of developers actually use.

🛠️ Available Tools (All Working)

📤 ExportToLLMButton Patterns

Static feeds:context=\"static\" staticPath=\"exports/spec\"
Current page:context=\"current\"
Mini clipboard:mini clipboard
Developer mode:showCurlCommand showDirectUrl

⚡ Implementation Reality

Minimal valid feed (2 minutes):

🌐 Basic MCP Feed - Copy & Customize🔍 Live
Replace {{placeholders}} with your actual values
{\n  \"feed_type\": \"mcp\",\n  \"metadata\": {\n    \"title\": \"{{Your Service Name}}\",\n    \"origin\": \"{{https://yoursite.com}}\"\n  },\n  \"intent\": \"data_processing\"\n}
That's it. LLMs understand this immediately - no training needed.

🔜 Coming Soon (Realistic Timeline)

Q3 2025:
  • CLI Tools (mcp validate, test, lint)
  • WordPress & Shopify plugins
  • GitHub Actions integration
Q4 2025:
  • VS Code & JetBrains plugins
  • SDK packages (JavaScript, Python, Go)
  • CI/CD templates
2026+:
  • Browser extension for feed detection
  • Enterprise analytics & monitoring
  • Industry-specific compliance tools

🎯 Why Implement Now?

  • First-mover advantage: Be ready when agents go mainstream
  • Low cost: Basic MCP feed takes 15 minutes
  • Future-proofing: Web standards take 3-5 years to mature
  • It's just JSON: No complex infrastructure needed
  • Works today: All major LLMs parse it natively

🧪 Test It Right Now

Kung Fu Test: Export our spec feed and ask any LLM \"What is LLMFeed?\"

NEW
← Try this, then paste into Claude/GPT

Ready to build with the agent web?

🥋 Start Here: Train Any LLM

Transform ChatGPT, Claude, or Gemini into MCP experts in 30 seconds. Then ask them to generate everything you need.

🧠 Get Training Prompt
Browse All Tools30-Second SetupJoin Community
Philosophy: Use LLMs for custom solutions, official tools for standardized workflows with proven traction.
", "markdown": "# Developer Tools & SDK\n\nEverything you need to build, validate, and deploy agent-ready feeds\n\n[X](https://twitter.com/intent/tweet?text=Discover%20this%20MCP-enabled%20resource&url=https%3A%2F%2Fwellknownmcp.org%2Fsdk \"Share on X\")[LinkedIn](https://www.linkedin.com/shareArticle?mini=true&url=https%3A%2F%2Fwellknownmcp.org%2Fsdk&title=Discover%20this%20MCP-enabled%20resource \"Share on LinkedIn\")[WhatsApp](https://wa.me/?text=Discover%20this%20MCP-enabled%20resource%20https%3A%2F%2Fwellknownmcp.org%2Fsdk \"Share on WhatsApp\")Copy link\n\nExport LLMFeedNEW\n\nComplete ecosystem of tools for creating, testing, and signing LLMFeed blocks. From visual builders to cryptographic verification - all the tools agents need.\n\n## 🎯 Current Status: Early but Functional\n\n**Honestly? Limited adoption.** We're in early experimental phase with dozens of feeds, not thousands. But the tech foundation is solid, and all major LLMs parse LLMFeed JSON natively.\n\n## 📦 The 3 Essential Feeds (WellKnownMCP Core)\n\n**spec.llmfeed.json** → Complete LLMFeed specification, LLM-optimized\n\n**compiled-site.llmfeed.json** → Entire website compiled & cleaned\n\n**news-export.llmfeed.json** → Full news archive with inline content\n\n**Usage:** Add any of these to your LLM's project knowledge for instant LLMFeed understanding.\n\n## 🧠 The Real Secret: Modern LLMs Are Already Perfect Tools\n\n**Honest truth:** You don't need our tools. Any modern LLM (ChatGPT, Claude, Gemini) can generate perfect LLMFeed JSON, write validation scripts, and create deployment automation.\n\n✅ What LLMs can do RIGHT NOW:\n\n* **Generate feeds:** Perfect mcp.llmfeed.json from your site description\n* **Write scripts:** Validation, signing, export automation in any language\n* **Create components:** React export buttons, Vue.js integrations, etc.\n* **Build workflows:** CI/CD pipelines, deployment automation\n* **Explain concepts:** Business ROI, technical architecture, compliance\n\n🥋\n\n**Train any LLM in 30 seconds:**[Copy our universal training prompt](/train)\n\nWorks with ChatGPT, Claude, Gemini - transforms them into MCP experts instantly\n\n### 🏗️ Official Tools Strategy: Traction-Driven Development\n\n**Our approach:** Build official tools based on actual developer adoption, not hypothetical needs.\n\n📈 High Traction → Official Tools\n\n* CLI packages (mcp validate, lint, watch)\n* IDE plugins (VS Code, JetBrains)\n* Platform integrations (WordPress, Shopify)\n* Enterprise SDKs (JavaScript, Python, Go)\n\n🧠 Always Available → LLM Generated\n\n* Custom validation scripts for your stack\n* One-off automation tools\n* Rapid prototyping and testing\n* Educational examples and tutorials\n\n**💡 Why this works:** LLMs excel at one-off custom solutions. Official tools make sense for repeated, standardized workflows that lots of developers actually use.\n\n## 🛠️ Available Tools (All Working)\n\n* **LLMFeedHub:**[Visual testing & agent simulation](/llmfeedhub)\n* **LLMFeedForge:**[Visual feed builder with one-click signing](https://llmfeedforge.org)\n* **Export Button:**[React component for feed generation](/tools/export-button)\n* **Schema Validation:**[Canonical, annotated, lite schemas](/tools/schema)\n* **Verification API:**[Cryptographic validation tools](/tools/sign-and-verify)\n* **LLMCA Certification:**[Third-party trust verification](https://llmca.org)Soon: delegated signing (LLMCA private key + user breadcrumbs)\n\n### 📤 ExportToLLMButton Patterns\n\n**Static feeds:**`context=\"static\" staticPath=\"exports/spec\"`\n\n**Current page:**`context=\"current\"`\n\n**Mini clipboard:**`mini clipboard`\n\n**Developer mode:**`showCurlCommand showDirectUrl`\n\n### ⚡ Implementation Reality\n\n**Minimal valid feed (2 minutes):**\n\n🌐 Basic MCP Feed - Copy & Customize🔍 Live\n\nReplace {{placeholders}} with your actual valuesCopy\n\n```\n{\n \"feed_type\": \"mcp\",\n \"metadata\": {\n \"title\": \"{{Your Service Name}}\",\n \"origin\": \"{{https://yoursite.com}}\"\n },\n \"intent\": \"data_processing\"\n}\n```\n\n**That's it.** LLMs understand this immediately - no training needed.\n\n## 🔜 Coming Soon (Realistic Timeline)\n\n**Q3 2025:**\n\n* CLI Tools (mcp validate, test, lint)\n* WordPress & Shopify plugins\n* GitHub Actions integration\n\n**Q4 2025:**\n\n* VS Code & JetBrains plugins\n* SDK packages (JavaScript, Python, Go)\n* CI/CD templates\n\n**2026+:**\n\n* Browser extension for feed detection\n* Enterprise analytics & monitoring\n* Industry-specific compliance tools\n\n### 🎯 Why Implement Now?\n\n* **First-mover advantage:** Be ready when agents go mainstream\n* **Low cost:** Basic MCP feed takes 15 minutes\n* **Future-proofing:** Web standards take 3-5 years to mature\n* **It's just JSON:** No complex infrastructure needed\n* **Works today:** All major LLMs parse it natively\n\n### 🧪 Test It Right Now\n\n**Kung Fu Test:** Export our spec feed and ask any LLM \"What is LLMFeed?\"\n\nExport LLMFeedNEW\n\n← Try this, then paste into Claude/GPT\n\nReady to build with the agent web?\n\n🥋 Start Here: Train Any LLM\n\nTransform ChatGPT, Claude, or Gemini into MCP experts in 30 seconds. Then ask them to generate everything you need.\n\n[🧠 Get Training Prompt](/train)\n\n[Browse All Tools](/tools)[30-Second Setup](/tools/well-known)[Join Community](/join)\n\n**Philosophy:** Use LLMs for custom solutions, official tools for standardized workflows with proven traction.", "title": "Developer Tools & SDK — Build & Sign LLMFeed", "word_count": 645, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of sdk for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Complete SDK documentation with libraries, code examples, and integration guides.", "tags": [ "sdk", "libraries", "development", "integration" ], "description_llm": "Comprehensive SDK docs with examples and guides", "intent": "develop", "importance": "critical", "audience": [ "developer", "software-engineer" ] }, { "block_name": "join", "type": "html+markdown", "timestamp": "2026-07-20T19:10:42.005797Z", "content": { "html": "

Interested in LLMFeed?

LLMFeed is an open standard for making websites agent-readable. Whether you're curious about the technology, want to implement it, or explore partnership opportunities — we'd love to hear from you.

📄

Read the Spec

Understand the standard

View spec →
🛠️

Try the Tools

Generate your first feed

Explore tools →
📰

Latest News

Updates & articles

Read news →

Direct contact: opensource@wellknownmcp.org

", "markdown": "# Interested in LLMFeed?\n\nLLMFeed is an open standard for making websites agent-readable. Whether you're curious about the technology, want to implement it, or explore partnership opportunities — we'd love to hear from you.\n\nName\n\nEmail\n\nWhat brings you here?Select an option...I want to implement LLMFeed on my siteI'm building an AI agent/toolResearch / Academic interestPartnership / Business opportunityI want to contribute to the projectJust curious about the technologyOther\n\nTell us more (optional)\n\nGet in touch\n\n📄\n\n### Read the Spec\n\nUnderstand the standard\n\n[View spec →](/spec)\n\n🛠️\n\n### Try the Tools\n\nGenerate your first feed\n\n[Explore tools →](/tools)\n\n📰\n\n### Latest News\n\nUpdates & articles\n\n[Read news →](/en/news)\n\nDirect contact: [opensource@wellknownmcp.org](mailto:opensource@wellknownmcp.org)", "title": "Join the WellKnownMCP Community", "word_count": 112, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of join for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Community onboarding with governance participation and contribution opportunities.", "tags": [ "community", "governance", "participation", "onboarding" ], "description_llm": "Entry point for LLMFeed ecosystem participation", "intent": "convert", "importance": "medium", "audience": [ "community", "contributor" ] }, { "block_name": "ecosystem", "type": "summary", "timestamp": "2026-07-20T19:10:42.031050Z", "content": { "title": "MCP Ecosystem Showcase — Pioneer Agent-Web Sites", "summary": "# MCP Ecosystem Showcase\nPioneer sites leading the transition to the Agent Web — Signed, Certified, and Future-Ready\nThese pioneering organizations are implementing the Model Context Protocol, creating the foundation for trustworthy agent-web interactions.\nPioneer NetworkTrust VerifiedOpen Ecosystem\n## MCP Pioneer Network", "word_count": 42, "content_type": "summary_extract", "note": "Content summarized due to length - full content available on site" }, "context_summary": { "origin_type": "summarized page content", "motivation": "Lightweight summary of ecosystem for agent overview", "recommended_agent_actions": [ "Get overview of key concepts", "Identify main topics and sections", "Direct users to full page for details" ] }, "llm_summary": "LLMFeed ecosystem overview including partners, integrations, and community projects.", "tags": [ "ecosystem", "partners", "integrations", "community" ], "description_llm": "Complete ecosystem landscape and partnership network", "intent": "explore", "importance": "medium", "audience": [ "business", "developer", "community" ] }, { "block_name": "feeds", "type": "summary", "timestamp": "2026-07-20T19:10:42.040765Z", "content": { "title": "LLMFeed Examples Collection — Agent-Ready Feeds", "summary": "Loading LLMFeed collection...", "word_count": 3, "content_type": "summary_extract", "note": "Content summarized due to length - full content available on site" }, "context_summary": { "origin_type": "summarized page content", "motivation": "Lightweight summary of feeds for agent overview", "recommended_agent_actions": [ "Get overview of key concepts", "Identify main topics and sections", "Direct users to full page for details" ] }, "llm_summary": "Searchable directory of public LLMFeed feeds with categorization and trust levels.", "tags": [ "directory", "catalog", "examples", "feeds" ], "description_llm": "Public LLMFeed feeds directory with examples and categories", "intent": "discover", "importance": "high", "audience": [ "developer", "implementer", "researcher" ] }, { "block_name": "verify", "type": "interface", "timestamp": "2026-07-20T19:10:42.051699Z", "content": { "title": "LLMFeed Signature Verification", "interface_type": "verification_tool", "capabilities": [ "Upload and verify MCP feed files", "Check cryptographic signatures", "Validate feed structure and compliance", "Display trust chain information" ], "instructions": [ "Upload a .llmfeed.json file or paste feed content", "Tool will verify signature and structure", "Results show trust level and any issues found" ], "content_type": "interactive_interface", "note": "This is an interactive tool - agents should direct users to use it directly" }, "context_summary": { "origin_type": "interactive web interface", "motivation": "Describe verify functionality for agent guidance", "recommended_agent_actions": [ "Explain tool capabilities to users", "Provide usage instructions", "Direct users to the interface URL", "Help interpret results if shared" ] }, "llm_summary": "Interactive tool for verifying LLMFeed feed signatures and validating trust chains.", "tags": [ "verification", "security", "validation", "trust" ], "description_llm": "Security-focused feed verification and signature validation", "intent": "verify", "importance": "critical", "audience": [ "security-engineer", "developer", "auditor" ] }, { "block_name": "llmfeedhub", "type": "interface", "timestamp": "2026-07-20T19:10:42.072877Z", "content": { "title": "LLMFeedHub — Universal Feed Analyzer", "interface_type": "testing_platform", "capabilities": [ "Preview and test MCP feeds", "Simulate agent interactions", "Validate feed formats and structure", "Test different feed types and scenarios" ], "instructions": [ "Upload feed file or enter feed URL", "Use simulation tools to test agent interactions", "Validate structure and content formatting" ], "content_type": "interactive_interface", "note": "This is an interactive tool - agents should direct users to use it directly" }, "context_summary": { "origin_type": "interactive web interface", "motivation": "Describe llmfeedhub functionality for agent guidance", "recommended_agent_actions": [ "Explain tool capabilities to users", "Provide usage instructions", "Direct users to the interface URL", "Help interpret results if shared" ] }, "llm_summary": "Interactive platform for previewing, testing, and analyzing LLMFeed feeds with simulation.", "tags": [ "platform", "simulation", "preview", "testing" ], "description_llm": "Comprehensive feed testing and agent simulation platform", "intent": "test", "importance": "high", "audience": [ "developer", "tester", "implementer" ] }, { "block_name": "llmfeedhub-preview", "type": "interface", "timestamp": "2026-07-20T19:10:42.118037Z", "content": { "title": "Agent Website Comparison — Generate Your Analysis Prompt", "interface_type": "analysis_tool", "capabilities": [ "Analyze any website URL for MCP compatibility", "Check for /.well-known/ directory structure", "Discover available feeds and capabilities", "Preview how agents would interpret the site" ], "instructions": [ "Enter any website URL", "Tool scans for MCP feeds and well-known structure", "Results show agent-readable content and trust level" ], "content_type": "interactive_interface", "note": "This is an interactive tool - agents should direct users to use it directly" }, "context_summary": { "origin_type": "interactive web interface", "motivation": "Describe llmfeedhub-preview functionality for agent guidance", "recommended_agent_actions": [ "Explain tool capabilities to users", "Provide usage instructions", "Direct users to the interface URL", "Help interpret results if shared" ] }, "llm_summary": "Website analysis tool inspecting sites for LLMFeed feeds and agent-readable content.", "tags": [ "preview", "discovery", "analysis", "well-known" ], "description_llm": "Website LLMFeed discovery and agent-readiness evaluation", "intent": "analyze", "importance": "high", "audience": [ "developer", "webmaster", "auditor" ] }, { "block_name": "legal", "type": "html+markdown", "timestamp": "2026-07-20T19:10:42.129685Z", "content": { "html": "

Legal notice

The Foundation

Site Publisher

This site is published by the non-profit organization (association loi 1901):
Association WellKnownMCP (in the process of creation)
Address: (official address forthcoming)
Email: contact@wellknownmcp.org

Publication Director

The publication director is the acting president of the association WellKnownMCP.

Hosting Provider

This site is hosted by:
OVH SAS
2 rue Kellermann, 59100 Roubaix, France
Phone: 1007

Personal Data Protection (GDPR)

This website does not collect personal data, except information necessary to process requests sent via provided email addresses. In accordance with the GDPR, you have the right to access, rectify, delete, and oppose your personal data. You can exercise these rights by sending an email to contact@wellknownmcp.org.

Cookies

This website does not use any tracking or analytical cookies.

Intellectual Property

All content available on this website, including the MCP specification, is published under the MIT License. Trademarks, logos, and other distinctive signs used on this site are protected under applicable intellectual property laws.

Certification and Signatures

The certification authority for official LLMFeed signatures is currently operated by LLMCA. Certification policies and the public verification key are available at llmca.org/.well-known/llmca_cert.pem.

Official Repository

The official repository of the specification is publicly accessible at: github.com/wellknownmcp/spec. It includes detailed specifications, changelogs, and signed reference files.

", "markdown": "# Legal notice\n\nThe Foundation\n\n## Site Publisher\n\nThis site is published by the non-profit organization (association loi 1901): \n**Association WellKnownMCP (in the process of creation)** \nAddress: (official address forthcoming) \nEmail: [contact@wellknownmcp.org](mailto:contact@wellknownmcp.org)\n\n## Publication Director\n\nThe publication director is the acting president of the association WellKnownMCP.\n\n## Hosting Provider\n\nThis site is hosted by: \n**OVH SAS** \n2 rue Kellermann, 59100 Roubaix, France \nPhone: 1007\n\n## Personal Data Protection (GDPR)\n\nThis website does not collect personal data, except information necessary to process requests sent via provided email addresses. In accordance with the GDPR, you have the right to access, rectify, delete, and oppose your personal data. You can exercise these rights by sending an email to [contact@wellknownmcp.org](mailto:contact@wellknownmcp.org).\n\n## Cookies\n\nThis website does not use any tracking or analytical cookies.\n\n## Intellectual Property\n\nAll content available on this website, including the MCP specification, is published under the MIT License. Trademarks, logos, and other distinctive signs used on this site are protected under applicable intellectual property laws.\n\n## Certification and Signatures\n\nThe certification authority for official LLMFeed signatures is currently operated by **LLMCA**. Certification policies and the public verification key are available at [llmca.org/.well-known/llmca\\_cert.pem](https://llmca.org/.well-known/llmca_cert.pem).\n\n## Official Repository\n\nThe official repository of the specification is publicly accessible at: [github.com/wellknownmcp/spec](https://github.com/wellknownmcp/spec). It includes detailed specifications, changelogs, and signed reference files.", "title": "Legal Notice | WellKnownMCP | WellKnownMCP", "word_count": 214, "content_type": "html+markdown" }, "context_summary": { "origin_type": "html+markdown page from build directory", "motivation": "Structured export of legal for agent consumption", "recommended_agent_actions": [ "Summarize content", "Extract key concepts", "Identify action items", "Cross-reference with other pages" ] }, "llm_summary": "Legal framework, governance structure, and certification authority policies.", "tags": [ "legal", "terms", "privacy", "governance" ], "description_llm": "Complete legal documentation and ecosystem governance", "intent": "inform", "importance": "medium", "audience": [ "legal", "compliance", "business" ] } ] } }