uv

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill provides the official installation commands for the uv tool, which involve fetching and executing scripts from astral.sh directly in the shell or PowerShell. Evidence includes the use of curl and sh for Linux/macOS and Invoke-RestMethod for Windows.
  • [COMMAND_EXECUTION]: Includes a shell script named init-project.sh that automates project setup by creating directories and files, and running uv commands to manage dependencies.
  • [PROMPT_INJECTION]: As a developer tool guide, the skill describes processes that ingest data from project configuration files like pyproject.toml and uv.lock, which creates a surface for indirect prompt injection if an agent processes untrusted project files. Evidence chain: Ingestion points are project files (pyproject.toml, uv.lock); Boundary markers are absent; Capability inventory includes shell execution via uv run and uv tool install; Sanitization of project configuration data is not implemented.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 04:17 PM
Security Audit — agent-trust-hub — uv