skills/camacho/ai-skills/bail/Gen Agent Trust Hub

bail

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes standard developer tools including git and the GitHub CLI (gh) to perform branch management, commits, and issue updates. These actions are aligned with the skill's stated purpose of cleaning up development environments.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection. 1. Ingestion point: User-provided 'reason' string in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Uses git and gh CLI commands to write to GitHub and local Git history as described in SKILL.md. 4. Sanitization: Absent; the input is directly interpolated into shell strings, which could allow for command injection if the execution environment does not provide its own layer of sanitization.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 08:15 PM