io.github.lu-zhengda/mcp-python-exec-sandbox

Python Exec Sandbox

Is the Python Exec Sandbox MCP server down? No. It answered initialize and tools/list in the probe of 2026-10-27.

Sandboxed Python execution for AI agents. PEP 723 inline deps, multi-version Python, zero pollution.

alive probed 2026-11-14 · unchanged since 2026-10-27

3tools

1.2ktokens in Claude Opus 5.5

0.6 %of Claude's 200k window

0.5×the median server

8.1 sto initialize

3 msto list tools

Run it: uvx mcp-python-exec-sandbox@0.1.5 (stdio, probed in a container with the network off)
Host: pypi.org
Reports itself as: mcp-python-exec-sandbox 2.14.7
Capabilities: experimental, prompts, resources, tasks, tools
Instructions: none
Of a 1M window: 0.1 % of a 1M context window
Count: 1.2k tokens in Claude Opus 5.5, exact, measured 2026-10-27. With the Claude Code prefix mcp__…__ on each tool name: 1.3k. The o200k_base estimate is 726. Counted again when the tool definitions change. Method.
Repository: https://github.com/lu-zhengda/mcp-python-exec-sandbox
Registry: active, updated 2026-02-09 · registry entry
Changes: 2 recorded, reachable in 2 of them. A row is added when the result, the tool count or the cost changes.

Most expensive tools

The Claude column scales the o200k estimate of each tool by this server's ratio of exact to estimate, 1.69×.

Tool	≈ Claude tokens	o200k estimate	Share of this server
execute_python	835	495	68 %
validate_script	307	182	25 %
check_environment	83	49	7 %

Other registry entries on pypi.org, the ones next to this name; all of them.

Servers that mention sandbox.

We add a row when the result changes. A server with no change keeps its last row.

Probed	Result	Init	Tools	Claude	o200k
2026-10-27 06:55	alive	2.3 s	3	1.2k	726
2026-10-20 06:43	alive	8.1 s	3		726

The result of the latest probe, the tool count and the token cost. The feed has an entry for every probe that changed them.

Stops answering, comes back, tool set or cost changes. One confirmation mail first, a stop link in every mail.