// SCAN REPORT

Python

14 tools. 9 can modify or destroy data without limits.

9 write tools that can modify data. Rate limits recommended.

Last updated: 31 Jul 2026

9 can modify or destroy data

5 read-only

14 tools total

Community server · catalogue entry checked 31/07/2026 · full schemas captured for 14 of 14 tools

How to control Python ↓

TOOL MAP

What Python exposes to your agents

Read (5) Write / Execute (9) Destructive / Financial (0)

Running more than one server? Check your whole stack →

CONTEXT-WINDOW COST

What Python costs in tokens

5,644 tokens of tool definitions, loaded on every request

2.8% of a 200k context window

689 heaviest tool: mypy

Full per-tool breakdown → Model it in the token calculator →

MOST DANGEROUS TOOLS

High Risk

The most dangerous Python tools

Execute · High conda The tool executes conda commands including environment creation, removal, and package updates. Execute · High poetry This tool executes Poetry commands including package installation, addition, removal, and build operations. Execute · High uv-run This tool executes arbitrary commands within a uv-managed Python environment. Execute · High uv-install Installing packages via uv pip install executes package setup scripts and arbitrary code from potentially untrusted sources.

9 of Python's 14 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Python

PolicyLayer is an MCP gateway — it sits between your AI agents and Python, and nothing reaches the server without passing your rules. These are the rules we recommend:

Cap read operations

{
  "black": {
    "limits": [
      {
        "counter": "black_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Python — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON PYTHON →

Instant setup, no code required.

FULL CATALOGUE

All 14 Python tools

EXECUTE 9 tools

Execute conda Runs conda commands (list, info, env-list, create, remove, update) and returns structured JSON output. Execute mypy Runs mypy and returns structured type-check diagnostics (file, line, severity, message, code). Execute pip-install Runs pip install and returns a structured summary of installed packages. WARNING: may execute untrusted code. Execute poetry Runs Poetry commands and returns structured output. Supports install, add, remove, show, build, update, lock, Execute pyenv Manages Python versions via pyenv. Execute pytest Runs pytest and returns structured test results (passed, failed, errors, skipped, failures). Execute ruff-check Runs ruff check and returns structured lint diagnostics (file, line, code, message). Execute uv-install Runs uv pip install and returns a structured summary of installed packages. WARNING: may execute untrusted cod Execute uv-run Runs a command in a uv-managed environment and returns structured output.

READ 5 tools

Read black Runs Black code formatter and returns structured results (files changed, unchanged, would reformat). Read pip-audit Runs pip-audit and returns a structured vulnerability report. Read pip-list Runs pip list and returns a structured list of installed packages. Read pip-show Runs pip show and returns structured package metadata (name, version, summary, dependencies). Supports multipl Read ruff-format Runs ruff format and returns structured results (files changed, file list).

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about Python

Is the Python MCP server safe to use without restrictions? +

The Python server is primarily read-only with 5 read tools. While it cannot modify data, an agent in a retry loop can make thousands of API calls per minute, exhausting rate limits and running up costs. Rate limiting is still recommended.

How many tools does the Python MCP server expose? +

14 tools across 4 categories: Destructive, Execute, Read, Write. 5 are read-only. 9 can modify, create, or delete data.

How do I enforce a policy on Python? +

Register the Python MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Python tool call.

Deterministic rules across all 14 Python tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON PYTHON →

Instant setup, no code required.

14 Python tools catalogued and risk-classified — across an index of 46,500+ MCP servers.

// WHERE THIS COMES FROM

These policies come from Python's registry record.

The record behind this page: verified identity, auth posture, risk grade, every tool classified, recommended policy — re-checked continuously.

Teams ship this data inside their own products. See what a licence covers →