Mathlas MCP Server

Airtight math for AI agents: 3.68M-doc theorem search + numeric/Lean verification. No LLM, no key. This MCP entry surfaces a curated ranking/list from the project README rather than a classic multi-tool integration surface. It runs locally over stdio via the published package.

People connecting documentation tools to Claude, Cursor, VS Code, or another MCP client. The project is written in Python.

VERIFIED ACTIVE

LAST COMMIT 2026-07-25 · ★ 11 · #74 OF 183 MAINTAINED DOCUMENTATION · VERIFIED 2026-08-25

Apache-2.0 · Python servers · how we verify → /methodology

01 · Install Mathlas

before you install - you'll need

The README does not document required environment variables for a basic install.

Claude Code

claude mcp add archerkattri-mathlas -- uvx mathlas-mcp

Claude Desktop / Cursor / VS Code - add to config

{
  "mcpServers": {
    "archerkattri-mathlas": {
      "command": "uvx",
      "args": [
        "mathlas-mcp"
      ]
    }
  }
}

Same JSON for Cursor. For VS Code, rename the top-level key from `mcpServers` to `servers`.

Using another client? Same JSON, different key

Claude Desktop · mcpServers

Cursor · mcpServers

VS Code · servers

Windsurf · mcpServers

Zed · context_servers

Cline · mcpServers

Roo Code · mcpServers

Continue · mcpServers

LibreChat · mcpServers

Gemini CLI · mcpServers

Codex CLI · mcp_servers

Full setup guides: every client.

02 · Evidence

Security posture

What to check before giving this server access to your agent - from the registry, GitHub, and our own probes. We don't score safety; we show what's verifiable.

runs as local process (stdio) - runs on your machine with your user's permissions

license Apache-2.0 - declared in the repository

pypi package mathlas-mcp - check the name against the project README before installing (PyPI has no namespace ownership)

registry namespace io.github.Archerkattri is GitHub-verified and matches the repo owner

03 · What Mathlas can do

Prose above is summarized from the project's README and registry record - no invented capabilities.

Latest releases

v1.5.0 · 2026-07-02

This release adds a reranker tier and the DeepMind formal-conjectures corpus, and moves the Lean toolchain to 4.31.0. · Highlights · jina-reranker-v3 optional reranker tier - opt in via MATHLAS_RERANK_MODEL for a…

v1.4.0 · 2026-06-11

What's new in v1.4.0 · Since v1.3.0: · MCP ToolAnnotations on every tool - all 12 tools now declare behaviour hints (readOnlyHint / idempotentHint / openWorldHint / destructiveHint) on both the FastMCP and the…

v1.3.0 · 2026-06-10

The 0.6B end-to-end laptop tier · The same 3,683,428-document corpus and served representation channel, re-embedded once with Qwen3-Embedding-0.6B (1024-d, row-aligned with the served meta), so the query encoder itself…

04 · Who maintains Mathlas

mathlas is maintained by archerkattri. It's the only MCP server we track from this author; the repo dates to Jun 2026.

05 · Facts

category
documentation - ranked #74 of 183 actively-maintained documentation servers as of 2026-08-25.
release cadence
7 releases in the last 90 days (latest 2026-07-02)
registry
io.github.Archerkattri/mathlas (active, first published 2026-06-09 · 6 versions)
packages
pypi:mathlas-mcp

06 · Mathlas FAQ

What is Mathlas?

Airtight math for AI agents: 3.68M-doc theorem search + numeric/Lean verification. No LLM, no key. This MCP entry surfaces a curated ranking/list from the project README rather than a classic multi-tool integration surface. It runs locally over stdio via the published package.

Is Mathlas still maintained?

Yes - as of 2026-08-25, its last commit was 2026-07-25 and it shipped 7 releases in the last 90 days. We re-verify nightly.

How do I install Mathlas?

Run `uvx mathlas-mcp`. The README does not document required environment variables for a basic install. You can also paste the ready-made client config above.

Does Mathlas run locally?

Yes - it's a stdio server: it runs on your machine (via uvx) with your user's permissions. Your data stays local unless the server itself calls external APIs.

07 · Alternatives to Mathlas