English Français العربية

cloud.fetcher/fetcher

Official registry Search and web

The web capability layer for AI agents: render, extract, DNS, SSL, WHOIS & more via x402.

Source code Website Documentation

Install

Add this to your client's MCP configuration (Claude Desktop: claude_desktop_config.json; Cursor: .cursor/mcp.json; VS Code and others use the same "mcpServers" shape).

Remote (streamable-http)
Claude Code 
claude mcp add --transport http fetcher 'https://api.fetcher.cloud/mcp'
JSON configuration (Claude Desktop, Cursor, VS Code) 
{
    "mcpServers": {
        "fetcher": {
            "type": "http",
            "url": "https://api.fetcher.cloud/mcp"
        }
    }
}

Documentation

From the project's README on GitHub, refreshed weekly. Links open on GitHub.

Contents
  1. fetcher.cloud — the web capability layer for AI agents
  2. Why it exists
  3. How agents use it
  4. The capabilities
  5. Machine-readable surfaces
  6. Repository layout
  7. Deploy
  8. License

fetcher.cloud — the web capability layer for AI agents

One MCP server and API that gives an autonomous AI agent the public web. Render and screenshot pages, extract clean content, resolve links, parse feeds, and inspect domains, DNS, SSL, and WHOIS — 18 capabilities, paid per-call in USDC via x402, with a free daily tier.

Whenever an agent needs something from the internet, it uses fetcher.

Why it exists

LLMs are brilliant at reasoning and blind at retrieval. The moment real-world data needs a browser, a payment, or a login, an agent stalls. fetcher is the specialist capability layer it calls — a single interface for everything an agent can't do alone, billed the only way that works for software: per call, in stablecoins, with no account or API key.

How agents use it

  1. Discover — GET /capabilities returns every capability with category, pricing, examples, and schemas. Or connect the MCP server at /mcp.
  2. Call — a normal GET request. The first unpaid call past the free tier returns 402 Payment Required with machine-readable terms.
  3. Pay — the agent's wallet pays a fraction of a cent in USDC on Base and retries. No human in the loop.

Works with Conway Automaton, Claude, OpenAI Agents, Google ADK, any MCP client, or any agent that can make an HTTP request.

The capabilities

CategoryCapabilities
Browser & renderingrender, screenshot
Content & extractionextract, unfurl, feed, oembed, pdf-text, diff
Web intelligenceresolve, headers, robots, favicon
Domain & DNSdomain, domain/enriched, whois, dns, ssl
Utilityqr

Full catalog with pricing and examples: https://api.fetcher.cloud/capabilities

Machine-readable surfaces

/ (manifest) · /capabilities · /pricing.json · /openapi.json · /mcp · /.well-known/ai-plugin.json · llms.txt

Repository layout

worker/         # the Cloudflare Worker (API, payments, MCP)   → see worker/README.md
website/        # the landing page (Cloudflare Pages)
test-client/    # a script that pays an endpoint on Base mainnet
conway-skill/   # ready-to-submit Conway Automaton skill

Deploy

See worker/README.md and DEPLOY.md for step-by-step deployment (Cloudflare Workers + Pages). Payments require Coinbase CDP facilitator keys set as Worker secrets.

License

MIT — see LICENSE.

Frequently asked questions

What is cloud.fetcher/fetcher?

The web capability layer for AI agents: render, extract, DNS, SSL, WHOIS & more via x402.

How do I install cloud.fetcher/fetcher?

It runs as a remote server. In Claude Code: claude mcp add --transport http fetcher 'https://api.fetcher.cloud/mcp'. For other clients, paste the JSON configuration from this page into their MCP settings.

Does cloud.fetcher/fetcher need an API key?

Its registry entry lists no required key or secret. Some features may still need an account with the service it connects to.

Is cloud.fetcher/fetcher open source?

Yes. Its code is on GitHub under the MIT license.

Who maintains cloud.fetcher/fetcher?

It is published by fetchercloud. AI.tn lists it but does not maintain or endorse it.

From the official MCP Registry. Listing is not an endorsement: review a server's code and permissions before giving it access to your data.