conventional-commits
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses several shell commands to inspect the local repository state.
- Evidence: Workflow steps 2, 3, and 4 in
SKILL.mdinvokegit diff,git branch,git log, andgit showto extract metadata and code changes. - [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted content from the repository (staged code and commit history) which could contain malicious instructions.
- Ingestion points: The skill reads the output of
git diff --stagedandgit logas primary inputs for its analysis inSKILL.md(Workflow steps 3 and 4). - Boundary markers: The skill lacks specific instructions for the agent to treat the repository content strictly as data or to ignore embedded prompts within the code comments or commit messages.
- Capability inventory: The skill utilizes shell-based command execution for data gathering.
- Sanitization: There is no evidence of filtering or sanitization performed on the content retrieved from the git repository before it is passed to the analysis logic.
Audit Metadata