Fetch URL
Hosted MCP server that fetches any URL and returns its readable plain text
Answered our latest tool-list check 6 hours ago with 1 tools. It will be marked verified after 3 successful checks in 7 days.
Fetch URL is an open-source MCP server for web search and scraping. It runs as a hosted remote server, so there is nothing to install locally.
- Pricing
- Open source
- Platforms
- API, Self-hosted
- Runs
- Remotely
- Version
- 0.0.3
- Listing updated
Similar MCP servers
Install
Run in a terminal
claude mcp add --transport http fetch-url https://mcp.build/mcp/fetch-urlAdd to Settings → Connectors → Add custom connector
https://mcp.build/mcp/fetch-urlAdd to ~/.cursor/mcp.json
{
"mcpServers": {
"fetch-url": {
"url": "https://mcp.build/mcp/fetch-url"
}
}
}Add to .vscode/mcp.json
{
"servers": {
"fetch-url": {
"type": "http",
"url": "https://mcp.build/mcp/fetch-url"
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"fetch-url": {
"serverUrl": "https://mcp.build/mcp/fetch-url"
}
}
}Add to Grok Bot desktop app → Connect apps → Add a custom connector
https://mcp.build/mcp/fetch-urlEnter a Name (for example "Fetch URL"), choose Type: Remote HTTPS, paste this as the Server URL and choose Add plugin.
Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"fetch-url": {
"type": "http",
"url": "https://mcp.build/mcp/fetch-url"
}
}
}Using several servers? Build one config file with the MCP config generator.
About Fetch URL
The server takes a URL from an AI client, makes a plain HTTP GET request with a standard User-Agent while following redirects, and strips scripts, styles, navigation and footer markup so the agent receives paragraph text. The README and site describe it as open source and self-hostable with a small Python app configured through FETCH_MAX_CHARS and FETCH_TIMEOUT.
No API key is needed to connect to the hosted endpoint, which makes it a quick way to give a model read access to public pages such as documentation. It is part of the mcp.build collection of Hal9-structured MCP servers that run on Cloudish.
It suits developers and agent builders who want an MCP client to read a public page without leaving the conversation. The listing does not describe support for pages that require login or heavy client-side rendering.
Tools this server provides (1)
Read directly from the server.
- chat
- Fetch and extract the text content of any URL. Hosted at https://cloudish.ai.
What it does
- Plain text extraction
- Returns page content as clean paragraph text with scripts, styles, navigation and footers removed.
- No API key needed
- The hosted endpoint can be connected to directly without credentials.
- Redirect following
- Requests follow redirects and send a standard User-Agent header.
- Configurable limits
- Self-hosted runs can set maximum characters and fetch timeout through environment variables.
- Streamable HTTP transport
- Exposed as an MCP server over Streamable HTTP for common MCP clients.
Good for
- Letting an assistant read API or product documentation during a chat
- Checking a web page before recommending it to a user
- Pulling the text of a public article into an agent workflow
- Running a self-hosted URL reader for a private agent setup
Pricing
Open source
Open source and self-hostable; the hosted endpoint is listed without a stated fee.
Prices change. Check mcp.build before you buy.
Fetch URL FAQ
Does Fetch URL require an API key?
The site says no API key is required; you connect the endpoint and it works immediately.
Can I run Fetch URL myself?
Yes. The README gives local steps using pip install and a Python app, and the source is in the hal9ai/mcp.build repository.
What settings can I change when self-hosting?
FETCH_MAX_CHARS defaults to 4000 and FETCH_TIMEOUT defaults to 15 seconds.
Is Fetch URL free?
Yes. Fetch URL is open source, so you can use and self-host it for free.
How do I add Fetch URL to Claude Code?
Run this command in your terminal: claude mcp add --transport http fetch-url https://mcp.build/mcp/fetch-url
What are the best alternatives to Fetch URL?
The closest alternatives to Fetch URL are webfetch.run, Page Fetch, FetchMCP API, 4fetch and Fetch MCP Server.
Where this comes from
- Description: written from the project's README, the project's website and its MCP Registry entry.
- Status and tools: checked automatically by AIPoweredBot on 21 November 2026.
- Install steps: from the official MCP Registry entry.