Share:
About
Fetch any URL and return cleaned page text or raw HTML. A minimal, dependency-free way to pull docs, blog posts, or API references into the conversation without configuring a full scraping stack.
Example tools
Illustrative tool names — the actual tool set is defined by the server at runtime.
web_search
Search the web and return top results
fetch_url
Fetch and return the contents of a URL
Screenshots
No screenshots yet.
Frequently Asked Questions
Common questions about Fetch MCP.
- What is the primary function of the Fetch MCP?
- The Fetch MCP is designed to retrieve content from any given URL and return either cleaned page text or raw HTML. This enables users to integrate web content directly into their workflows without complex scraping configurations.
- How can I deploy the Fetch MCP with a client like Cline?
- The Fetch MCP uses the stdio transport. For Cline, you would download the server from the official repository and configure Cline to run the server executable on demand, passing the URL as an argument through standard input.
- Are there any authentication requirements for using the Fetch MCP?
- No, the Fetch MCP does not require an API key, OAuth, or any other credentials for operation. It is designed for straightforward, unauthenticated access to public web content.
- What specific capabilities does the Fetch MCP offer?
- The Fetch MCP provides the core functionality to fetch web pages, offering options to return either human-readable cleaned text or the raw HTML source. It focuses on simplicity and dependency-free operation.
- Does the Fetch MCP incur any costs for its use?
- The Fetch MCP itself is open-source and free to use. Any operational costs would depend on your specific deployment environment and how you choose to host and run the server, such as compute or network egress charges.
- What is a key limitation of the Fetch MCP regarding content retrieval?
- A key limitation is that the Fetch MCP primarily retrieves static content. It may not fully render or interact with dynamic content generated by client-side JavaScript, potentially returning incomplete information for highly interactive web pages.
Install Fetch MCP
Claude Desktop
Add this to claude_desktop_config.json.
{
"mcpServers": {
"fetch": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-fetch"
]
}
}
}Cursor
Add this to ~/.cursor/mcp.json.
{
"mcpServers": {
"fetch": {
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-fetch"
]
}
}
}VS Code
Add this to your workspace settings.json.
{
"mcp.servers": {
"fetch": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@modelcontextprotocol/server-fetch"
]
}
}
}Reviews
Sign in to leave a review.
No reviews yet. Be the first!
Discussion(0)
No comments yet.