Atlas / MCP servers / yzfly / Python Interpreter

Python InterpreterBLOCK

mcp/yzfly/python-interpreter

MCP Python Interpreter: run python code. Python-mcp-server, mcp-python-server, Code Executor

Verdict
BLOCK
Grade
D
Trust score
69 /100
Exposed tools
10 5r · 4w · 1d
Transport
stdio · streamable-http
License
NOASSERTION
Stars
104
01

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

[](https://mseep.ai/app/yzfly-mcp-python-interpreter)

A Model Context Protocol (MCP) server that allows LLMs to interact with Python environments, read and write files, execute Python code, and manage development workflows.

Features

  • Environment Management: List and use different Python environments (system and conda)
  • Code Execution: Run Python code or scripts in any available environment
  • Package Management: List installed packages and install new ones
  • File Operations:
  • Read files of any type (text, source code, binary)
  • Write text and binary files
  • Python Prompts: Templates for common Python tasks like function creation and debugging

Installation

You can install the MCP Python Interpreter using pip:

pip install mcp-python-interpreter

Or with uv:

uv install mcp-python-interpreter

Usage with Claude Desktop

  1. Install Claude Desktop
  2. Open Claude Desktop, click on menu, then Settings
  3. Go to Developer tab and click "Edit Config"
  4. Add the following to your claude_desktop_config.json:
{
"mcpServers": {
"mcp-python-interpreter": {
"command": "uvx",
"args": [
"mcp-python-interpreter",
"--dir",
"/path/to/your/work/dir",
"--python-path",
"/path/to/your/python"
],
"env": {
"MCP_ALLOW_SYSTEM_ACCESS": 0
},
}
}
}

For Windows:

{
"mcpServers": {
"python-interpreter": {
"command": "uvx",
"args": [
"mcp-python-interpreter",
"--dir",
"C:\\path\\to\\your\\working\\directory",
"--python-path",
"/path/to/your/python"
],
"env": {
"MCP_ALLOW_SYSTEM_ACCESS": "0"
},
}
}
}
  1. Restart Claude Desktop
  2. You should
Read from source at commit 858b5910b5f0OBSERVED · 2026-10-07
02

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.

claude-code
claude mcp add mcp-python-interpreter -- uvx mcp-python-interpreter
claude-desktop
{
  "mcpServers": {
    "mcp-python-interpreter": {
      "command": "uvx",
      "args": [
        "mcp-python-interpreter"
      ]
    }
  }
}
03

Exposed tools (10)

5 read · 4 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.

ToolRiskDescription
clear_sessiondestructive
install_packagewrite
list_directoryread
list_installed_packagesread
list_python_environmentsreadList all available Python environments (system Python and conda environments).
list_sessionsreadList all active REPL sessions.
read_fileread
run_python_codewrite
run_python_filewrite
write_filewrite
04

Trust audit

BLOCKgrade D · trust 69/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.

LayerWhat it checksResult
L0Provenance & inventoryPASS
L1Static analysis of the codeFAIL
L2Instruction surface (what it tells the agent)PASS
L3Class-specific surfaceWARN
L4Behavioural (sandbox)SKIPPED

What the source does

Filesystem
declared (1 observation(s))
Network
none-observed
Shell
declared (2 observation(s))
Dependencies
pinned
Secrets in source
none-found

Findings (4)

HIGHCode injection · code.eval_exec · CWE-78, CWE-94, CWE-95
mcp_python_interpreter/server.py:97
result_value = eval(code, self.locals)
Why it matters. evaluates text as code
Fix. remove; use a parser or a dispatch table
HIGHCode injection · code.eval_exec · CWE-78, CWE-94, CWE-95
mcp_python_interpreter/server.py:102
exec(code, self.locals)
Why it matters. evaluates text as code
Fix. remove; use a parser or a dispatch table
MEDIUMFilesystem / path · mcp.destructive_tools · CWE-22, CWE-59
clear_session
Why it matters. 1 tool(s) can delete or overwrite
Fix. prefer a read-only mode or scoped tokens; the page states the blast radius
INFOSupply chain · prompt.pipe_to_shell · CWE-829, CWE-1357
README.md:89
curl -LsSf https://astral.sh/uv/install.sh | sh

Gates applied: no_behavioural_pass.

Audited 2026-10-07 · audit v0.4.1 · source sha 858b5910b5f0full audit observations/trust-audit/mcp-server/yzfly__python-interpreter.json · Report an issue / request a re-scan
05

Audit history

Every audit this server has had. A grade with a past is a grade somebody is still checking.

DateSourceVerdictGradeScoreChange
2026-10-07858b5910b5f0BLOCKD69first audit
06

Questions

What is the Python Interpreter MCP server?

MCP Python Interpreter: run python code. Python-mcp-server, mcp-python-server, Code Executor

What tools does Python Interpreter expose?

10 in total: 5 read-only, 4 that write, and 1 that can delete or overwrite (clear_session). Every one is listed on this page with its risk.

Is Python Interpreter safe to connect to an agent?

No — not without reading the findings first. The audit graded it D (69/100) and found 2 critical or high issues in the source. Each one is listed on this page with the file and line it is on. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.

What credentials does Python Interpreter need?

No credential environment variables were found in its source, so it appears to need none.

How does Python Interpreter run?

It speaks stdio and streamable-http, so it runs as a local process your client starts. It is published on PyPI as mcp-python-interpreter.

How current is this page?

The grade is for one exact copy of the source (858b5910b5f0), read on 2026-10-07. The repository is watched and re-audited when it changes.

Advertisement