Python Exec Sandbox
This server gives Claude the ability to execute Python code in throwaway sandboxes that vanish after each run. Scripts declare dependencies inline using PEP 723 metadata blocks, uv resolves them ephemerally, and nothing touches your host Python installation. It exposes execute_python for running code with configurable timeouts, validate_script for checking metadata without execution, and check_environment for inspecting the runtime. On Linux it uses bubblewrap for namespace isolation, elsewhere it spins up Docker containers. You can target Python 3.13 through 3.15 regardless of what's on your host machine. Reach for this when you need Claude to prototype with third-party packages or analyze data without polluting your system with accumulating virtualenvs and package conflicts.
- tools
- 3
- transport
- STDIO
- last updated
- Feb 9, 2026
Tools
3 toolsPublic tool metadata for what this MCP can expose to an agent.
execute_pythonExecute a Python script with automatic dependency management.
Execute a Python script with automatic dependency management.
No parameter schema in public metadata yet.
check_environmentReturns information about the execution environment: Python version, uv version, platform, sandbox status, and configuration.
Returns information about the execution environment: Python version, uv version, platform, sandbox status, and configuration.
No parameter schema in public metadata yet.
validate_scriptValidates a script's PEP 723 metadata and dependencies without executing it.
Validates a script's PEP 723 metadata and dependencies without executing it.
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageReadme
mcp-python-exec-sandbox
Sandboxed Python execution for AI agents. Scripts run in ephemeral, isolated environments with inline dependencies (PEP 723) -- zero host pollution, zero leftover venvs, zero package conflicts.
Why?
Every coding agent can already run Python on your host. The problem is what happens next: packages accumulate, venvs sprawl, and a rogue pip install breaks your system. mcp-python-exec-sandbox eliminates this:
Details
- Repository
- github.com/lu-zhengda/mcp-python-exec-sandbox
- Package
mcp-python-exec-sandbox- Transport
- STDIO
- Registry
- MCP registry, active