A Model Context Server providing simple features to automate python script execution via LLMs
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"python-interpreter-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
## python-interpreter-mcp: A MCP server to run scripts ## Overview A lightweight, experimental MCP server designed to execute arbitrary Python scripts in a structured and reproducible environment. It leverages **[uv](https://github.com/astral-sh/uv)** to run isolated code snippets through subprocesses. ### Tools - `run_script` - Runs the given script with `uv run script.py`. - Input: - `code`(str): Script to be run - Return: The stdout of the given script. ## Configuration ### Usage with OpenAI Agents SDK ```python async with MCPServerStdio( params={ "command": "uvx", "args": ["python-interpreter-mcp"], } ) as server ... ``` ### Usage with Claude Desktop Add this to your `claude_desktop_config.json`: ```json "mcpServers": { "interpreter": { "command": "uvx", "args": ["python-interpreter-mcp"] } } ``` ## How it works - A script string is received by the MCP tool `run_script`. - A hidden folder is created in cwd, and the script is saved as a `.py` file inside it. - The script is then executed using `uv run`, which ensures dependency isolation. - The stdout of the script is captured and returned as the response. ## Usage Warnings **This project is in a very early stage of development.** ### ⚠️ Important notes - It executes arbitrary Python code, which means it can run anything — including malicious or destructive commands. - Use only in trusted, sandboxed environments. - You should always validate, guardrail, or restrict inputs when wiring this into an LLM.
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.