--- title: "Fetch MCP reference server: review, setup & limitations · Undominated.ai" canonical: https://undominated.ai/mcp-servers/fetch/ description: "Retrieve a URL as text or Markdown in chunks through a small local MCP server." --- # Fetch MCP reference server: review, setup & limitations · Undominated.ai > Retrieve a URL as text or Markdown in chunks through a small local MCP server. [← Explore all mcp servers](/mcp-servers/) WEB RESEARCH / Model Context Protocol steering group # Fetch MCP reference server Retrieve a URL as text or Markdown in chunks through a small local MCP server. See setup guidance ↓Original source ↗ SOURCE REVIEW Reviewed 2026-09-21 Evidence 5 linked sources Publisher Model Context Protocol steering group Licence Apache-2.0 / MIT transition; documentation CC-BY-4.0 ↗ Revision d73f99efbfd4 Read what was—and wasn’t—checked ↓ “Web content fetching and conversion for efficient LLM usage.” Model Context Protocol steering group · upstream description ↗ Our analysis follows below. 01 / THE REASONING ## Why this made the selection. - Provides direct URL retrieval without a search-service API account. - Chunked extraction and optional raw responses make long pages usable by a client. ### A good fit for - Fetch a known public documentation page. - Prototype a client that reads long web responses in parts. ### Weigh up before choosing - The publisher describes this as an educational reference implementation, not a production-ready service. - The server can reach local or internal IP addresses; restrict its network environment before exposing it to untrusted requests. - This is an HTTP fetcher, not a browser that renders JavaScript or a general web search index. - Robots handling differs between autonomous tool calls and user-initiated prompts; it is not a universal access-policy enforcement layer. 02 / THE REVIEW RECORD ## What we actually inspected. Source review has boundaries. A clear record is more useful than a “safe” badge. ### Material inspected - src/fetch/README.md, pyproject.toml and server.py. - Robots checks, HTTP request handling, extraction and stdio startup. - Repository README and current licence transition. ### Our findings - Autonomous fetch calls check robots.txt by default. The user-prompt path follows a separate code path. - The package exposes response offsets and raw output; no browser renderer or hosted search backend is configured. ### Not established by this review - Server launch, authenticated tool calls, and release-to-source parity were not tested. The review applies to the material and revision named here. A newer upstream release can change its behavior. 03 / PUT IT TO WORK ## Connect a server deliberately. Upstream setup instructions ↗ DOCUMENTED COMMAND pip install mcp-server-fetch Copy command ↗ Copying does not execute this command. It may retrieve a newer version than the reviewed source. - Install the package with pip, or configure the documented uvx subprocess in the MCP client. - Use the Claude Desktop JSON example below; other clients use different configuration keys. - Place the process in an appropriate network environment and only request sources you are allowed to access. ### Before you start - Python tooling (pip or uv). - Outbound access to the requested sites; optional Node.js HTML simplification. ### Documented configuration example Use the client and authentication context described in the setup guidance. Configuration formats vary between hosts. { "mcpServers": { "fetch": { "command": "uvx", "args": [ "mcp-server-fetch" ] } } } Copy configuration ↗ ### Compatibility Claude Desktop · stdio clients ### Implementation Python ### Transports stdio ### Local process - Makes HTTP requests from the machine running the server, including to internal addresses reachable by that machine. - Returns retrieved content to the connected MCP client. ### Cost model Runs on your machine; hosting and any connected model client are separate costs. 04 / FOLLOW THE EVIDENCE ## The source trail. Our notes are separate from the original resource. Check upstream before adopting a new version. - Fetch MCP reference server setup and scope ↗ Checked 2026-09-21 https://raw.githubusercontent.com/modelcontextprotocol/servers/d73f99efbfd40c3aa1b61e88728b3d49fb52608f/src/fetch/README.md Supports: summary, upstreamDescription, whySelected, bestFor, limitations, install, access, compatibility, transports, review - MCP steering-group reference implementation scope ↗ Checked 2026-09-21 https://raw.githubusercontent.com/modelcontextprotocol/servers/d73f99efbfd40c3aa1b61e88728b3d49fb52608f/README.md Supports: publisher, upstreamDescription, limitations, review - Current MCP repository licence transition ↗ Checked 2026-09-21 https://raw.githubusercontent.com/modelcontextprotocol/servers/d73f99efbfd40c3aa1b61e88728b3d49fb52608f/LICENSE Supports: license, review - Fetch Python package ↗ Checked 2026-09-21 https://raw.githubusercontent.com/modelcontextprotocol/servers/d73f99efbfd40c3aa1b61e88728b3d49fb52608f/src/fetch/pyproject.toml Supports: language, install - Fetch HTTP and robots implementation ↗ Checked 2026-09-21 https://raw.githubusercontent.com/modelcontextprotocol/servers/d73f99efbfd40c3aa1b61e88728b3d49fb52608f/src/fetch/src/mcp_server_fetch/server.py Supports: review, limitations, access, transports KEEP COMPARING ## Other approaches to consider. Related by category or shared topics. These are alternatives to inspect, not a measured quality order. [### Exa MCP ↗ Connects an assistant to Exa web search, content retrieval and optional multi-step research tools.](/mcp-servers/exa/)[### Tavily MCP ↗ Combines web search, extraction, mapping and crawling through local stdio or Tavily’s hosted MCP service.](/mcp-servers/tavily/)[### Filesystem MCP reference server ↗ Give a local MCP client scoped file reading, editing, search, and directory operations.](/mcp-servers/filesystem/) [AI Tools ↗](/tools/)[Skills ↗](/skills/)[Agents ↗](/agents/)[MCP Servers ↗](/mcp-servers/)