uv-run
Runs a command in a uv-managed environment and returns structured output.
This record as markdown: /tools/io-github-dave-london-pare-python/uv-run.md
What uv-run does on Python
AI agents invoke uv-run to trigger actions in Python. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call: builds kicked off, notifications sent, workflows started.
| Parameter | Type | Required | Description |
|---|---|---|---|
path | string | — | Working directory |
frozen | boolean | — | Run without updating the lockfile (--frozen) |
locked | boolean | — | Assert the lockfile is up-to-date (--locked) |
module | boolean | — | Run as python -m <module> instead of a script (-m) |
noSync | boolean | — | Skip syncing the environment for faster execution (--no-sync) |
python | string | — | Python interpreter version to use (-p VERSION) |
command | array | Yes | Command and arguments to run (e.g. ['python', 'script.py']) |
compact | boolean | — | Prefer compact output |
envFile | string | — | Path to environment file (--env-file FILE) |
isolated | boolean | — | Run in a clean isolated environment (--isolated) |
outputLimit | integer | — | Maximum characters to return for stdout/stderr before truncation |
withPackages | array | — | Additional packages to inject for one-off use (--with PKG) |
Parameters from the server's own tool schema.
Why uv-run is rated High
This tool executes arbitrary commands within a uv-managed Python environment. While uv is a package manager, the ability to 'run a command' means an AI agent could execute potentially harmful operations including system commands, data exfiltration, or lateral movement.
From the tool's definition Tool description explicitly states 'Runs a command in a uv-managed environment', indicating arbitrary command execution capability. The verb 'runs' combined with 'command' in an environment context strongly indicates Execute category.
Risk signalsAccepts freeform code/query input (python) · Accepts file system path (path) · High parameter count (12 properties)
Attacks that exploit this kind of access
The rule that runs uv-run safely
PolicyLayer is an MCP gateway: it sits between your AI agents and Python, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For uv-run, this is the rule to start with:
uv-run stays usable, but rate-capped: a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect Python, apply this rule, and every uv-run call is checked against it from then on.
Questions about uv-run
Runs a command in a uv-managed environment and returns structured output. It is categorised as a Execute tool in the Python MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
uv-run accepts 12 parameters: path, frozen, locked, module, noSync, python, command, compact, envFile, isolated, outputLimit, withPackages. Required: command. The full parameter table on this page comes from the server's own tool schema.
Register the Python MCP server in PolicyLayer and add a rule for uv-run: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Python. Nothing to install.
uv-run is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the uv-run rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for uv-run. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
uv-run is provided by the Python MCP server (Dave-London/Pare). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on Python, and thousands of servers like it.
Across the catalogue