semver-changelog
Fail
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: HIGHPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- Indirect Prompt Injection (HIGH): The skill processes untrusted data from git commit logs and diffs to update project documentation, creating a significant injection surface.\n
- Ingestion points:
git log <latest_tag>..HEAD --onelineandgit diff <latest_tag>..HEAD(SKILL.md).\n - Boundary markers: Absent. There are no instructions or delimiters to isolate untrusted git output.\n
- Capability inventory: Authorized to perform file write operations on
CHANGELOG.md(SKILL.md).\n - Sanitization: Absent. No logic is provided to filter or validate commit messages.\n- Command Execution (LOW): The skill relies on the execution of local
gitcommands.\n - Evidence: Uses
git tag,git diff, andgit logto extract repository metadata. While standard, these are executed in the local environment.
Recommendations
- AI detected serious security threats
Audit Metadata