commit-helper
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to executegit difffor reviewing changes andgit commitfor applying the generated messages. It also executes a local Python script (scripts/validate_commit.py) to ensure messages adhere to formatting standards. - [PROMPT_INJECTION]: The skill processes untrusted data in the form of
git diffoutput. This presents a surface for indirect prompt injection where malicious instructions embedded in the codebase being committed could potentially influence the agent's behavior. However, the skill's capabilities are limited to Git operations and message formatting. - Ingestion points:
SKILL.mddescribes a workflow that involves runninggit diffto review changes. - Boundary markers: No specific delimiters or instructions to ignore embedded commands are present for the diff content.
- Capability inventory:
Bash,Read,Write,Edit,Greptools allowed inSKILL.md. - Sanitization: The
validate_commit.pyscript validates the structure of the message but does not sanitize the content of the diff itself.
Audit Metadata