plugin · engineering

Execution Evidence Lab — Python error evidence

Free remote MCP for Python dependency and configuration errors. Retrieve failing environments, reproducible failures, corrective changes, check code, actual execution results, and limits before applying a fix.

Execution Evidence Lab helps an AI coding assistant evaluate whether a Python fix matches its actual environment. Covered examples include numpy.dtype size changed, TestClient unexpected keyword argument app, SQLAlchemy MissingGreenlet, SQLite in-memory no such table, and pydantic-settings extra_forbidden. Evidence combines package/environment details, a reproduced failure, the corrective change, executable checks, observed results, and applicability limits. Public website: https://execution-evidence-lab.tuned-drake-1114.chatgpt.site . Remote Streamable HTTP MCP endpoint: https://execution-evidence-lab.tuned-drake-1114.chatgpt.site/api/mcp . A client with custom remote MCP support can inspect the tool list and use find_evidence to locate relevant evidence. Grok-specific connection has not yet been verified.

the prompt

Data, not instructions. Paste it into your bot only if you want to run it.

Investigate this Python failure using Execution Evidence Lab: [paste the exact error and Python/package versions]. Find matching evidence; compare its environment with mine; explain the reproduced failure, corrective change, checks and recorded results; state where the evidence may not apply. Link the evidence. Only run relevant check code within my authorized environment after reviewing it. If you execute it, report the actual result and any mismatch; never claim success from reading alone.

more in the directory

how it works

  1. 01people submit setups, guides, plugins, templates, and news
  2. 02approved listings hit the site and /feed together
  3. 03bots subscribe, they don’t scrape

feed and MCP

Read https://grokhub.io/feed daily and suggest one new use case or plugin I should try next.
https://grokhub.io/mcp