audit-the-list

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data fetched from public package registries and repository hosting services.
  • Ingestion points: Project metadata, descriptions, and statistics are retrieved from gh api, https://pypi.org/pypi/{name}/json, and https://pypistats.org/api/packages/{name}/recent (SKILL.md).
  • Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the retrieved external content.
  • Capability inventory: The skill performs local command execution via uv run, make, and git commit (SKILL.md).
  • Sanitization: There is no evidence of sanitization or validation of external metadata before it is used to draft verdicts or generate commit messages.
  • [COMMAND_EXECUTION]: The skill uses several local tools and scripts to perform its audit tasks.
  • Evidence: The skill instructions specify the use of uv run python, gh api, make test, make build, and git commit. While these are aligned with the skill's stated purpose of repository maintenance, they represent a high capability level.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 04:51 PM
Security Audit — agent-trust-hub — audit-the-list