# Context Relay

> URL-to-Markdown web extraction and context compression for AI agents. Fetch a public web page and return clean, task-relevant, citation-verifiable LLM context with exact excerpts, locators, hashes, and estimated tokens saved. Use before an agent summarizes, answers questions from, or cites a long or noisy URL. Deterministic and model-free. Public HTML only; no PDFs, logins, or JavaScript-only pages.

- Capabilities: https://context-relay.contextrelay.workers.dev/api/v1/capabilities
- OpenAPI: https://context-relay.contextrelay.workers.dev/openapi.json
- Free quote: POST https://context-relay.contextrelay.workers.dev/api/v1/context/quote
- Paid x402 retrieval: POST https://context-relay.contextrelay.workers.dev/api/v1/context
- MCP Streamable HTTP: https://context-relay.contextrelay.workers.dev/mcp (tools: quote_context, retrieve_context)
- Health: https://context-relay.contextrelay.workers.dev/api/v1/health
- Price: $0.005 USDC on eip155:8453
- Inputs: url, optional question, max_tokens, max_price_usd, freshness_seconds
- Output: versioned Markdown context, exact excerpts, SHA-256 hashes, safety signals, unit-economics estimate, and delivery receipt
- Settlement proof: PAYMENT-RESPONSE header; failed handler responses are not settled
- Scope: public HTML only; robots enforced; no login, paywall, browser automation, PDF, or private-network access