Habr MCP server
habr-mcp
An MCP server for searching and reading Habr articles: converts an article's markup into readable markdown right in the agent's reply
- 0 ★ on the repository
- Last commit
Extended access
Why this level
- Runs a Python server that searches and reads articles through the internal habr.com API
Install
- Works in
- Services
Manual install
pip install httpx mcp && python habr_mcp.pyCommand assembled from the code's dependencies and entry point; the repository has no separate install guide.
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Runs a Python server that searches and reads articles through the internal habr.com API
This is third-party code. Review the repository files before installing.
What it does
A FastMCP server calls Habr's internal API and gives the agent two actions: search articles by query with relevance sorting in Russian, and read a specific article in full. The article's HTML markup (headers, lists, code, links, bold, italics) is converted to markdown by its own small converter with no external parsing library. It is a single file, depending only on httpx and mcp.
Who it is for. For developers who want to search and read Habr articles from a conversation with an agent without opening the site.
Good fit when
- You want to quickly find Habr articles on a topic and get them as markdown
- You want an agent to read a specific article and reference it in its answer
Not a fit when
- You need article comments or author profile data: the server does not return these
- You need Habr's official documented API: this uses an undocumented internal endpoint
Example request
Find recent Habr articles about MCP servers and show me the content of the most relevant oneMCP
- Transport
- stdio
- Authentication
- not required
Limitations
It relies on Habr's undocumented internal habr.com/kek/v2 endpoint, which can change without notice. The HTML-to-markdown converter is minimal and may lose some formatting on unusual layouts. There is no README in the repository; purpose and usage are visible only from the code.
How to disable. Remove the habr server from your MCP client config.
FAQ
Do I need a Habr account?
No, the server uses public search and article reading without authorization.
Can it read non-Russian articles?
Search defaults to fl=ru and hl=ru; you would need to change those parameters in the code for another language.
Discussion
No questions or reviews yet.
Related
All Habr toolsHabr MCP server for reading and publishing
Habr MCP server
A multi-tenant HTTP MCP server for Habr: anonymous article reading and publishing drafts from Google Docs or Docmost
Habr article parser MCP server
habr_mcp
A Dockerized MCP server that pulls a Habr article's text by link, aimed at local models in LM Studio
Habr Career MCP server
habr-career-mcp
An MCP server for the Habr Career API: filtered vacancy search, full vacancy text, a skills lookup and a local registry of vacancies already reviewed
Habr Career MCP server
habr-mcp
An MCP server that edits your own career.habr.com profile using a session cookie, because the official API is read-only
MCP server that reads sites on your machine and hands the agent short Markdown with links, keeping raw HTML out of context.
A web search MCP server that gives the agent full page text at once: Stack Overflow answers, GitHub threads, arXiv papers
README in short
The repository has no README: the single-file habr_mcp.py shows a FastMCP server with two tools, habr_search and an article reader, plus its own HTML-to-markdown converter supporting headers, lists, links, code and paragraphs.