review-and-commit-msg
Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands including
git status,git diff,git add, andgit commit. These operations interact directly with the local repository and filesystem to automate the development workflow. - [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection (Category 8) because it processes untrusted data from source code changes. Ingestion points: Untrusted data enters the agent context through the
WBS modal textinput and the output ofgit diff. Boundary markers: There are no explicit delimiters used to encapsulate the diff content or instructions to the LLM to ignore embedded commands within the code. Capability inventory: The skill has the capability to modify the repository state usinggit addandgit commit. Sanitization: No validation or sanitization is performed on the diff content or WBS text before it is analyzed by the agent. A partial mitigation is present in Step 6, which requires user selection and confirmation before a commit is executed.
Audit Metadata