Start here

Get started with the spec

We know how you behave nowadays 🧠: you want to try before you read. Fair enough — just feed your favorite LLM with these three .llmfeed.json files 🥢. It's already a good start.
Yes, these buttons exist in copy-to-clipboard mode too, but for education purposes they open the file first. Just read, copy, and feed. 🥋 Welcome to the dojo — if your LLM says "I know Kung-fu", it worked.
Not sure what this is? Read the spec ->
"Just give me a well-formed .llmfeed.json, and I'll do the rest."
🌐

Website in a capsule

Includes the main pages of the site, cleaned and compiled. Best for a quick LLM overview of the entire project.

✓ Signed✓ Agent-optimized✓ Complete context
NEW
Open JSON
Curl Command
curl -s wellknownmcp.org/.well-known/exports/compiled-site.llmfeed.json | jq .
📚

Spec in a capsule

Complete LLMFeed specification from GitHub, optimized for LLM consumption. Perfect for understanding the technical standard.

✓ Complete specification✓ Implementation guide✓ Incremental builds
NEW
Open JSON
Curl Command
curl -s wellknownmcp.org/.well-known/exports/spec.llmfeed.json | jq .
📰

News archive (EN)

Complete archive of all WellKnownMCP news articles with full content inline. Generated incrementally at build time for optimal performance.

⚡ Build-time static⚡ Full content⚡ Incremental updates
NEW
Open JSON
Curl Command
curl -s wellknownmcp.org/.well-known/exports/news-export.llmfeed.json | jq .

🥋 Test Your Agent Understanding

Download our feeds below and ask your LLM about them. If it says "I know Kung Fu" after reading them, you've successfully parsed the protocol!

  1. 1.Copy any feed URL above
  2. 2.Paste it into Claude or ChatGPT
  3. 3.Ask: "Do you know Kung Fu?"
  4. 4.If it says yes, it worked! 🥋

A correct response indicates successful protocol parsing and demonstrates the agent understands MCP structure.

Quick test command:
curl -s wellknownmcp.org/.well-known/mcp.llmfeed.json | jq -r '.prompts[]? | select(.keywords[]? | contains("kung fu")) | .intent'
Expected output: "easter egg intro to the full site"

🔗 Quick Access for Agents

Direct links to core MCP infrastructure files. Perfect for agent discovery and validation. Includes JSON schemas for feed validation and development.

/.well-known/mcp.llmfeed.jsonMain MCP Feed
/.well-known/llm-index.llmfeed.jsonFeed Index
/.well-known/capabilities.llmfeed.jsonCapabilities
/.well-known/manifesto.llmfeed.jsonManifesto
/.well-known/schema.llmfeed.jsonJSON Schema
/.well-known/schema.annotated.llmfeed.jsonSchema (Annotated)
🚀 Agent Power User Command (All Exports):
for export in compiled-site spec news-export; do echo "=== $export ===" && curl -s wellknownmcp.org/.well-known/exports/$export.llmfeed.json | jq -r '.metadata.title // .title'; done
Downloads all major static exports with full content
📋 Feed Validation (JSON Schema):
curl -s wellknownmcp.org/.well-known/mcp.llmfeed.json -> feed.json && curl -s wellknownmcp.org/.well-known/schema.llmfeed.json -> schema.json && ajv validate -s schema.json -d feed.json
Download and validate any feed against the official JSON schema (requires ajv-cli)

💡 Why Structured Feeds Work Better

❌ Traditional Scraping

  • AI agents scraping HTML
  • High token costs
  • Slow responses
  • Unreliable parsing
  • No verification

✅ LLMFeed Solution

  • Structured JSON feeds
  • 96.7% fewer tokens
  • 10x faster responses
  • Reliable data structure
  • Cryptographic signatures
Result:Same content, optimized delivery. Humans get interactive websites, agents get structured data.