uv
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions for downloading and installing Python packages from PyPI using commands like
uv add,uv pip install, anduvx. It includes a specific warning regarding the use ofuvxwith unknown tools. - [COMMAND_EXECUTION]: The skill documentation encourages the execution of local scripts and command-line tools through
uv runanduvxas part of the standard Python development workflow. - [INDIRECT_PROMPT_INJECTION]: The skill processes project metadata and lockfiles which could theoretically contain malicious instructions if a project is sourced from an untrusted origin.
- Ingestion points: Local configuration files including
pyproject.toml,uv.lock, andrequirements.txt(referenced in SKILL.md). - Boundary markers: The skill does not define specific delimiters for separating project data from agent instructions.
- Capability inventory: The skill uses
uvto execute code (uv run), install packages (uv add), and run CLI tools (uvx) across various scripts. - Sanitization: No specific sanitization or validation of the contents of
pyproject.tomloruv.lockis mentioned, though the guide advises users to only run well-known tools.
Audit Metadata