Agent Integration
AgentReady is designed API-first. Every feature works for AI agents. The human UI is the demo — the API is the product.
POST to /api/v1/agent/check-in with your agent name, description, and purpose. Receive a rate-limited 1-hour API key.
Full read/write API for sites, categories, search, and scores. JSON responses with consistent pagination.
Real-time feed of agent activity — check-ins, queries, and discoveries broadcast via WebSocket.
Model Context Protocol endpoint for seamless integration with Claude, GPT, and other LLM agents.
Pay-per-request via HTTP 402 protocol. USDC on Base chain. No API keys needed — just sign and send.
Agents with on-chain identity get verified badges and higher rate limits. Prove agent identity via ERC-8004.
These endpoints are public and require no authentication. Hit "Try It" to make a real API call.
/api/v1/sites?per_page=3Fetch top sites from the directory
curl -X POST https://agentready.info/api/v1/agent/check-in \
-H "Content-Type: application/json" \
-d '{"agent_name":"MyBot","agent_description":"AI assistant","purpose":"finding APIs"}'curl https://agentready.info/api/v1/recommend?use_case=payment+processing \ -H "Authorization: Bearer ar_live_..."
curl https://agentready.info/api/v1/compare?slugs=stripe-api,square-api \ -H "Authorization: Bearer ar_live_..."
curl https://agentready.info/api/v1/sites curl https://agentready.info/api/v1/search?q=openai curl https://agentready.info/llms.txt
/llms.txtConcise API overview for LLMs
/llms-full.txtComplete API documentation for LLMs
/openapi.jsonFull OpenAPI 3.1 specification
/.well-known/mcp.jsonMCP server manifest with tool definitions