skills/accolver/skill-maker/pr-qa/Gen Agent Trust Hub

pr-qa

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it reads and acts upon external data from pull request comments and local repository documentation without sanitization or explicit boundary markers.
  • Ingestion points: Retrieves pull request comments via the GitHub API and scans the repository for architectural documentation (e.g., ARCHITECTURE.md, README.md).
  • Boundary markers: None identified; the agent is instructed to directly incorporate findings from these external sources into code fixes or responses.
  • Capability inventory: Includes the ability to push code, modify pull requests, and execute commands using the GitHub CLI and security audit tools.
  • Sanitization: None identified.
  • [EXTERNAL_DOWNLOADS]: The skill suggests installing the well-known security utility gitleaks via standard package managers (npm, pnpm, bun, or brew) if the tool is missing from the environment.
  • [COMMAND_EXECUTION]: The workflow executes automated repository management and security auditing tasks using git, the GitHub CLI (gh), and several ecosystem-specific dependency auditors such as npm audit and pip audit.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 04:25 PM
Security Audit — agent-trust-hub — pr-qa