release
Pass
Audited by Gen Agent Trust Hub on Jun 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill performs extensive shell command execution using Git and package managers (npm, pnpm, yarn, cargo) to manage the release lifecycle. These actions are fundamental to the skill's purpose and are guarded by numerous exit conditions and manual confirmation steps.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection in Phase 4 when processing untrusted data from the git history.
- Ingestion points: Commit subjects, bodies, and SHAs are read from
git logto generate release notes. - Boundary markers: No explicit delimiters or instructions are used to isolate the git log output from the agent's internal reasoning.
- Capability inventory: The skill can perform file system modifications (manifest and lockfile edits) and Git operations including
git commit,git tag, andgit push. - Sanitization: There is no evidence of filtering or sanitization of commit messages before they are processed by the LLM for grouping and summary generation.
Audit Metadata