setup-site-skills

Pass

Audited by Gen Agent Trust Hub on Aug 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to run git remote -v to determine if the repository is hosted on GitHub, which helps decide between using the GitHub CLI or local markdown files for issue tracking. This is a standard and safe operation for project scaffolding.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted data from local project files to generate AI instructions.
  • Ingestion points: The skill reads package.json, pnpm-workspace.yaml, and various configuration files (astro.config.*, next.config.*, etc.) in the SKILL.md explore phase.
  • Boundary markers: There are no explicit delimiters or warnings to ignore embedded instructions within the processed project files.
  • Capability inventory: The skill has the capability to write and update markdown files, specifically CLAUDE.md and AGENTS.md, which serve as instruction sources for the agent.
  • Sanitization: No specific sanitization or validation logic is provided for the external content before it is interpolated into the generated documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 5, 2026, 02:13 AM
Security Audit — agent-trust-hub — setup-site-skills