pr-descriptions
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION] (SAFE): Employs standard GitHub CLI (gh) commands for repository management. These commands are executed locally using the user's authenticated session.
- [DATA_EXFILTRATION] (SAFE): Interactions are limited to the official GitHub API. No evidence of sensitive file access or unauthorized data transmission was found.
- [INDIRECT_PROMPT_INJECTION] (LOW): The skill creates a surface for writing pull request descriptions. 1. Ingestion points: The 'new body content' placeholder in SKILL.md represents the data entry point. 2. Boundary markers: The use of shell heredocs (<<'EOF') is recommended, which provides shell-level protection but does not filter the content itself. 3. Capability inventory: Includes 'gh api' and 'gh pr' calls across the documentation. 4. Sanitization: The skill does not implement content sanitization, relying on the agent or user to provide safe PR body text.
Audit Metadata