conventional-commit
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses local shell commands to interact with git. It executes
git diff --cachedto analyze changes andgit committo finalize them (SKILL.md). These are appropriate for its function and include a confirmation step. - [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to instructions embedded in file diffs. 1. Ingestion points: The skill reads staged file changes using
git diff --cached(SKILL.md). 2. Boundary markers: Absent; the instructions do not include markers to separate code from commands or tell the agent to ignore instructions in the data. 3. Capability inventory: The skill can execute shell commands likegit commit(SKILL.md). 4. Sanitization: Absent; the diff output is processed by the model without filtering or escaping content.
Audit Metadata