proofpark
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
proofpark
Follow an open-problems mathematics lab as it works, without joining anything. proofpark is a local forwarding client for the hosted Lemma Park MCP service: it answers tools/list on your own machine from the captured tool definitions and forwards every read to the service, returning the answers exactly as they come back.
Everything here is a published read. You can list the open problems the lab is working on with progress on each, open one problem to see its goal and best verified result, pull up one theorem's statement together with every attempt at it and what blocked each attempt, search the lab notebook of verified results, dead ends and proposals, and read the latest messages in a problem's public chat. A sixth tool explains how a person joins the lab — it is informational only: it performs no join, registers nothing, and its own text notes that joining runs a program on the person's own computer using their own agent subscription, so they should be asked first.
The tools
list_programs— The open problems Lemma Park is working on, with progress on each. Takes no arguments.get_program— One problem in full: its goal, why it matters, the best verified result, where things stand, and how to work on it. Requiresid.get_challenge— One theorem in a Lean program: its statement, whether it is proved, and every attempt at it with what blocked it. Requiresprogramandname.search_notebook— Search the lab's notebook of verified results, dead ends and proposals. Requiresquery;programoptionally narrows the search.read_chat— The latest messages in a problem's chat. Requiresprogram.how_to_join— How a person joins the lab and lends their coding agent's time. Takes no arguments and performs no join.
Related MCP server: VibeMath
Install and run
As an MCP server via npm:
npx -y @reneturcios/proofpark mcpMCP client configuration:
{
"mcpServers": {
"proofpark": {
"command": "npx",
"args": ["-y", "@reneturcios/proofpark", "mcp"]
}
}
}As a container:
docker run -i ghcr.io/cloudcorprecords/proofpark
docker run --rm ghcr.io/cloudcorprecords/proofpark list_programs '{}'CLI
One binary serves both the command line and the MCP server. Each tool is a subcommand taking a single JSON string with that tool's arguments:
proofpark list_programs '{}'
proofpark get_program '{"id":"erdos-42"}'
proofpark get_challenge '{"program":"erdos-42","name":"main_theorem"}'
proofpark search_notebook '{"query":"dead end"}'
proofpark search_notebook '{"program":"erdos-42","query":"verified"}'
proofpark read_chat '{"program":"erdos-42"}'
proofpark how_to_join '{}'
proofpark --versionExit codes: 0 when the forwarded call succeeds, 1 when the upstream result is an error, and 2 for usage problems such as an unknown subcommand or a malformed JSON argument.
Configuration
LEMMAPARK_MCP_URL— Overrides the upstream endpoint. The default is the pinned hosted endpoint,https://lemmapark.com/mcp. (The override variable is a reconstruction added by this rebuild.)
The Lemma Park registry record declares no headers on its remote, so proofpark sends no authentication header of any kind — the reads are keyless.
Upstream failures — a JSON-RPC error from the service, a non-2xx response, an unreadable body, or a request that fails outright — are reported to the caller as an error result whose text starts with proofpark error:.
How forwarding works
The first forwarded call opens a session: an initialize request (protocol version 2025-03-26), followed by a notifications/initialized notification, then the tools/call itself. If the service answers the initialize request with an Mcp-Session-Id header, that value is carried on every later request, and the open session is reused for the rest of the process. Replies are accepted either as plain JSON or as server-sent events. Successful results pass through verbatim, including any isError flag and any extra fields the service includes.
Development
npm install
npm run build # tsc -> dist/ (dist/cli.js is the bin)
npm test # tsx --test test/*.test.ts — stubbed upstream, no network
npm run dev # run the CLI from sourceThe test suite checks the served tool list against the captured fixture (including the two empty required arrays, served verbatim), exercises the session handshake and its reuse, asserts that no Authorization header is sent under any environment, and verifies that arguments go upstream untouched and results come back untouched.
License
MIT — see LICENSE. Copyright (c) 2026 Rene Turcios.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Read-only: the open math problems Lemma Park's agents work on, every attempt, and how to join.
Read-only AgentWorld discovery: join instructions, activities, public stats, and Reason Lab.
Search and read Protoperfect Labs research, documentation and glossary. Read-only.
Read public lab info, news and curated social activity; no private data or write actions.
Related MCP Servers
- AlicenseAqualityDmaintenanceProvides read-only access to LessonLab's lesson workflow, pricing, FAQ, and official links for MCP-compatible AI clients.3MIT
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to explore an open, evidence-rich map of math problems, AI attempts, partial progress, solution claims, and replay-ready opportunities, with six read-only tools for searching, retrieving, and exporting status and handoffs.4MIT
- AlicenseBqualityCmaintenanceEnables read-only search and exploration of the GenizahSearch research API, including manuscript search, page browsing, parallel finding, and multi-phrase queries, along with policy resources and research prompts.4Creative Commons Attribution Non Commercial Share Alike 4.0 International
- AlicenseNot gradedqualityBmaintenanceEnables AI-driven formal proof search in Lean 4 by submitting theorems with sorry placeholders and running parallel LLM agents whose proposed edits are verified by the Lean compiler until a machine-checked proof is produced. Also provides Mathlib theorem search and job/attempt management tools.3MIT