release
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on standard git and gh CLI commands to perform its core functions. All potentially destructive or public actions, such as pushing tags or creating releases, are protected by explicit user confirmation steps throughout the protocol (SKILL.md, Steps 5 and 6).
- [INDIRECT_PROMPT_INJECTION]: The skill ingests commit messages and PR titles from the repository to decide on versioning and draft release notes (SKILL.md, Step 2 and 3). This represents an attack surface where untrusted data could influence agent behavior. The skill mitigates this risk by requiring the user to review and approve the computed version and the drafted notes before any action is taken.
- [SAFE]: The skill includes explicit instructions to refuse operation on dirty or diverged repositories and strictly forbids re-pointing existing tags, which prevents accidental disruption of the repository state.
Audit Metadata