Deepwiki MCP
Deepwiki MCP Server
MCP server that fetches deepwiki.com pages, cleans them and returns Markdown to the agent, as one document or split by page
- 1,392 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Reaches the network and downloads deepwiki.com pages
- Runs a local Node.js process
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add mcp/deepwiki-mcp-serverDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Reaches the network and downloads deepwiki.com pages
- Runs a local Node.js process
This is third-party code. Review the repository files before installing.
What it does
The server takes a deepwiki.com link, crawls the related pages and returns their content as Markdown. It only accepts the deepwiki.com domain, strips headers, footers, navigation, scripts and ads, and rewrites links for Markdown. You can pick an output mode, a single document or a list of pages, and set crawl depth and concurrency. The tool is named deepwiki_fetch and any MCP client can call it. Note: per the author's warning the server currently does not work because DeepWiki closed scraping.
Who it is for. For developers who want to pull a repository's deepwiki.com documentation into text for an agent.
Good fit when
- You want to feed a library's deepwiki.com documentation to an agent
- You need a single Markdown document assembled from several deepwiki pages
Not a fit when
- The source is not on deepwiki.com; the server rejects other domains
- You need a maintained option; the author recommends the official DeepWiki MCP
Example request
Fetch the docs from deepwiki.com/shadcn-ui/ui and assemble them into a single MarkdownMCP
- Transport
- stdio, http
- Authentication
- not required
Environment variables
DEEPWIKI_MAX_CONCURRENCY- Maximum concurrent requests, default 5.
DEEPWIKI_REQUEST_TIMEOUT- Request timeout in milliseconds, default 30000.
DEEPWIKI_MAX_RETRIES- Retry attempts for failed requests, default 3.
DEEPWIKI_RETRY_DELAY- Base retry backoff delay in milliseconds, default 250.
Limitations
Per the notice at the top of the README the server currently does not work: DeepWiki closed scraping, and the author recommends the official DeepWiki MCP as a replacement. Only deepwiki.com URLs are processed. This is an unofficial project. Behavior is tuned via environment variables: concurrency, timeout, retry count and retry delay.
How to disable. Remove the mcp-deepwiki block from your MCP client configuration and restart it.
FAQ
Is this the official DeepWiki server?
No, it is an unofficial project. The author recommends the official DeepWiki MCP as a replacement.
Why do requests fail?
Per the author, DeepWiki closed scraping, so the server currently does not work.
Discussion
No questions or reviews yet.
Related
MCP serversRef's MCP server: documentation search for APIs and libraries that hands the agent only the relevant parts of pages
A Go crawler turns documentation sites into Markdown and gives agents offline search_docs and page reading over MCP.
Docs MCP Server
Grounded Docs MCP Server
A local MCP server that indexes docs from websites, GitHub, npm, PyPI and files and gives the agent version-aware search
MCP server that fetches the exact documentation section instead of the whole file: a heading index and byte-precise reads
A remote MCP server that gives the agent docs and code of any public GitHub repository with nothing to install
Deep Graph MCP: a code graph of a repository for agents
Deep Graph MCP
An MCP server that gives an agent a repository code graph: search by meaning, symbol links and the places an edit would touch
Comparison with similar tools
Cards that share the main search query, and how they differ.
| Tool | Official | Authentication | Repository stars | Main difference |
|---|---|---|---|---|
| Deepwiki MCP (this card) | no | not required | 1,392 | |
| Better Context (btca) | no | 1,151 | the btca-local skill makes an agent clone library repositories and | |
| RepoWiki | no | 306 | The CLI builds wikis from local or GitHub codebases, links modules and traces answers to source-file line ranges. | |
| OpenWiki | no | 17,914 | OpenWiki builds linked Markdown repository wikis, retains source evidence and lets agents search documentation across projects. |
README in short
The README shows how to wire the server via npx in Cursor and other MCP clients and documents the deepwiki_fetch tool with url, mode and maxDepth parameters. It lists response shapes: single document, page list, error and partial success. There is a local run from source, a pnpm build and a Docker option, plus environment variables for concurrency, timeout and retries. The top of the README warns that the server currently does not work because of scraping blocks on DeepWiki's side. MIT licensed.