automate-project-maintenance

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local development and formatting tools including rumdl, mise, nox, and biome to perform project maintenance tasks.
  • [INDIRECT_PROMPT_INJECTION]: The skill analyzes repository configuration files and source code to determine the appropriate maintenance workflow.
  • Ingestion points: Repository inspection for Python signals (pyproject.toml) and TypeScript signals (package.json) in SKILL.md.
  • Boundary markers: None present; the skill determines its branch of execution based on the presence of these files.
  • Capability inventory: Execution of shell commands (rumdl, mise, nox, biome) and secondary skill calls ($write-python-tests, $sync-github-repo-metadata) across SKILL.md.
  • Sanitization: No explicit sanitization of repository content is mentioned before it influences the branch detection logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 02:06 PM
Security Audit — agent-trust-hub — automate-project-maintenance