commit-helper
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
Bashtool to executegit diffto analyze changes andgit committo finalize messages. These operations are essential for the skill's functionality and are designed to be triggered after user approval. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) because it processes untrusted data (code changes) through
git diffto generate commit messages. - Ingestion points: Untrusted content from files in the repository enters the agent's context via the
git diffcommand as specified in the Workflow section ofSKILL.md. - Boundary markers: No explicit boundary markers or "ignore instructions" warnings are defined to separate the diff content from the agent's internal instructions.
- Capability inventory: The skill is granted access to
Bash,Read,Write, andEdittools in theSKILL.mdmetadata. - Sanitization: There is no evidence of sanitization, escaping, or filtering of the diff output before it is used for commit message generation.
Audit Metadata