release
Pass
Audited by Gen Agent Trust Hub on Jun 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
git,gh(GitHub CLI), andjqto manage repository state and interact with the GitHub API.\n - Evidence: Phases 1 through 5 utilize shell commands to fetch tags, analyze logs, query PR status, and push new release tags.\n- [PROMPT_INJECTION]: The skill exposes an indirect prompt injection surface by ingesting untrusted data from the repository history.\n
- Ingestion points: Commit messages (
git log) and PR metadata (gh pr list) are read from the repository (SKILL.md, Phase 2 & 3).\n - Boundary markers: The instructions do not specify delimiters or warnings to the model to separate commit data from internal instructions.\n
- Capability inventory: The skill has the ability to execute shell commands (
git tag,git push,gh release create) and write to the local filesystem (CHANGELOG.md).\n - Sanitization: No evidence of filtering or escaping commit content before it is interpolated into the natural language generation step.
Audit Metadata