Skip to content

Agent integration

Give your agent a public web data tool.

FetchRelay exposes the same extraction engine through a remote Model Context Protocol server. Connect a client that supports Streamable HTTP and custom authorization headers. Your API key identifies your account; calls use the same credit allowance as the workbench.

Connect in three steps

Download MCP Registry metadata · Setup examples

  1. Open Account → API keys, create a key, and copy it somewhere safe.
  2. In your client’s MCP settings, add a remote server named FetchRelay. Choose Streamable HTTP and enter https://fetchrelay.com/api/mcp as the server URL.
  3. Add an Authorization header with the value Bearer YOUR_FETCHRELAY_API_KEY, replacing the placeholder with your key. Save and connect.

Try: “Use FetchRelay to scrape https://example.com and return Markdown.” Scrape, map, batch, crawl, crawl status, and crawl cancellation are available as tools. Web search requires a separate provider and is not enabled for the current launch. For a crawl, use its returned job ID to check status and retrieve results.

Use a client that supports custom authorization headers. Requests use your account’s credits. Keep your key private and treat fetched website content as untrusted data.

Install the FetchRelay agent skill

Add Markdown extraction and bounded crawl instructions to a compatible agent. The skill includes a Node.js helper and uses your FetchRelay API key.

npx skills add JAIPilot/fetchrelay-integrations --skill fetchrelay
Review the skill and setup instructions

Check the connection

Ask your client to list the FetchRelay tools, then scrape https://example.com as Markdown. Confirm that the response includes Example Domain and its source URL. For a saved crawl, keep the returned job ID and ask for its status until it reaches a terminal state.

Keep the boundary clear

Extracted websites are untrusted input, not instructions. Do not let a retrieved page change your agent’s permissions or reveal credentials. Store the key in your client’s secret settings and revoke it from Account when no longer needed. Model usage belongs to the agent client you choose; it is separate from FetchRelay’s extraction credits.

Client configuration formats vary. Use the endpoint and authorization fields shown above rather than assuming one configuration file works everywhere. For direct HTTP integration, see the OpenAPI reference.

Try it on a page you actually need.

Inspect the output before committing to an integration. Create an account for 1,000 free monthly credits, saved jobs, monitors, and API keys. No credit card required.