commit-work

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill orchestrates local git operations and allows for the execution of project-defined verification scripts such as unit tests or linters.
  • [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection via the content of file diffs it processes, which is inherent to its code-review functionality. 1. Ingestion points: git status and git diff outputs in SKILL.md. 2. Boundary markers: No specific delimiters are used to separate diff content from instructions. 3. Capability inventory: git operations and local script execution. 4. Sanitization: None. Note: The skill explicitly directs the agent to check for hardcoded secrets and accidental debug code, which is a positive security practice.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 04:05 PM
Security Audit — agent-trust-hub — commit-work