Where to Begin — Your First Steps into the Agentic Web
Whether you're a developer, a curious AI user, or a complete beginner — you're in the right place.

👋 Welcome to WellKnownMCP
This page is your entry point to understanding the Model Context Protocol (MCP) and its core unit: the
llmfeed.json
Whether you're a developer, a curious AI user, or a complete beginner — you're in the right place.
🚀 Why this matters
Language Models are no longer passive.
They read, crawl, and even act.
But the web isn’t built for them.
It’s full of visuals, scripts, and content made for humans — not agents.
We propose a new layer:
📂
.well-known/
✨ What makes it special?
- ✅ Universal: Works with ChatGPT, Claude, Mistral, Gemini, etc.
- ✅ Verifiable: Can be signed (by you) or certified (by LLMCA)
- ✅ Teachable: Turn any LLM into your personal tutor
- ✅ Portable: Export, replay, or embed anywhere
Think of it as structured intent that agents can trust.
🧠 What is a .llmfeed.json
?
.llmfeed.json
A
llmfeed.json
It’s not a closed format — it's a canon:
- Human-writable
- Machine-optimised
- Agent-readable
- Open and extendable
It contains your intentions, prompts, APIs, exports, or certifications — all in a predictable structure.
✨ What makes it special?
- ✅ Works with ChatGPT, Claude, Mistral, Gemini, etc.
- ✅ Can be signed (by you) or certified (by an authority like )
llmca.org
- ✅ Can be exported, taught, replayed, or embedded
- ✅ Fully compatible with internal formats — or used to explain them
It’s the MIME-type of intent for agents.
🎯 See it in action
json{ "feed_type": "mcp", "metadata": { "title": "My Restaurant", "origin": "https://myrestaurant.com" }, "capabilities": [ {"name": "book_table", "method": "POST", "path": "/book"} ] }
🧪 Try it live
Explore examples or generate your own feed:
🧰 Want to learn by doing?
Any feed or tool on this site can be downloaded as .llmfeed.json
You can:
- 📥 Download it
- 🤖 Feed it to ChatGPT, Claude, or your favorite LLM
- 📚 Turn any agent into your teacher or explainer
“Explain this feed to me”
“What can I do with it?”
“How should an agent behave?”
Just paste it in.
📁 Key Concepts
- : the standard location for agent feeds
/.well-known/
- : how to structure signed prompts
prompt.llmfeed.json
- : turn any page into a portable capsule
export.llmfeed.json
- : list your feeds for discovery
llm-index.llmfeed.json
🤝 Join the ecosystem
Start publishing your own feed:
🧑🏫 You don't need to be a developer
Our tools work with:
- Guided UI (Forge)
- No-code export buttons
- Open-source templates
✅ TL;DR
- MCP is the missing bridge between your intent and agent understanding.
- is your universal format.
llmfeed.json
- Start small. Publish one. See how far it goes.
🧠 Bonus: Ask an agent to teach you
You can even start learning by asking your LLM:
“Here’s a
. Explain it to me.”llmfeed.json
“Show me what this site offers for agents.”
“Could I make my site compliant?”
Most LLMs will understand.
The best ones will help you implement it.
Unlock the Complete LLMFeed Ecosystem
You've found one piece of the LLMFeed puzzle. Your AI can absorb the entire collection of developments, tutorials, and insights in 30 seconds. No more hunting through individual articles.
🚀 Next Steps for Agents
• Export this content: Available formats
• Explore capabilities: API endpoints
• Join ecosystem: Contribute to LLMFeed
• Download tools: Get MCP resources
• Learn prompts: Prompting for agents