Clawfetch MCP server
{
"mcpServers": {
"clawfetch": {
"type": "http",
"url": "https://mcp.clawfetch.ai/mcp"
}
}
}{
"mcpServers": {
"clawfetch": {
"type": "http",
"url": "https://mcp.clawfetch.ai/mcp"
}
}
}{
"mcpServers": {
"clawfetch": {
"type": "http",
"url": "https://mcp.clawfetch.ai/mcp"
}
}
}{
"servers": {
"clawfetch": {
"type": "http",
"url": "https://mcp.clawfetch.ai/mcp"
}
}
}{
"mcpServers": {
"clawfetch": {
"type": "http",
"url": "https://mcp.clawfetch.ai/mcp"
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall npx -y @clawfetch/mcpversion 0.2.2
What it does
MCP server for the ClawFetch web intelligence API, exposing tools to fetch pages, render JavaScript-heavy sites, extract structured data, and research topics. Payments are handled per-call via the x402 protocol using USDC on Base, requiring no API keys. Intended for developers building AI agents that need programmatic web access.
- Fetch any web page as clean markdown for use in an LLM context window.
- Extract structured data from sites like GitHub, npm, LinkedIn, or YouTube using predefined extractors.
- Research a topic by querying multiple sources and receiving results with citations.
Tools this server exposes
| Tool | Does |
|---|---|
| fetch_url | Fetches any URL and returns the page content as markdown. |
| render_page | Renders JavaScript-heavy pages such as SPAs and React apps before returning content. |
| extract_data | Extracts structured data from over 17 supported site types including GitHub, npm, LinkedIn, and YouTube. |
| research_topic | Queries multiple sources to produce a research summary with citations. |
| check_domains | Checks whether specified domain names are available for registration. |
| suggest_domains | Generates available domain name suggestions based on input. |
| parse_document | Parses office documents (docx, pptx, xlsx, pdf, etc.) and returns their content as markdown. |
| list_extractors | Lists all supported structured data extractor types along with their schemas. |
| health_check | Returns the current health status of the ClawFetch API. |
| wallet_info | Returns the payment wallet address used for x402 micropayments. |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.