l0-test

Pass

Audited by Gen Agent Trust Hub on Jul 22, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill executes a script downloaded from a remote source to install development tools. * Evidence: The script runs curl -LsSf "https://astral.sh/uv/${UV_VERSION}/install.sh" | sh inside a container to install the uv package manager. * Context: The download targets a well-known service (Astral) and occurs within an isolated Docker environment.
  • [EXTERNAL_DOWNLOADS]: The skill retrieves software and environment configurations from external registries. * Evidence: It pulls a PyTorch Docker image from the NVIDIA Container Registry (nvcr.io) and downloads the uv installer from astral.sh.
  • [COMMAND_EXECUTION]: The skill uses local CLI tools to manage the development environment and run tests. * Evidence: It executes git commands (e.g., git diff, git ls-files) to identify changed files, docker run to execute the test suite, and uv to manage Python dependencies and linting.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 22, 2026, 07:48 AM
Security Audit — agent-trust-hub — l0-test