release
Pass
Audited by Gen Agent Trust Hub on May 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes multiple shell commands to manage Git state, interact with the GitHub CLI (
gh), and run various package managers such asnpm,bun,uv, andcargo.- [EXTERNAL_DOWNLOADS]: Fetches project dependencies and clones external repositories from GitHub to synchronize release artifacts and Homebrew formulas. These operations target well-known and trusted service infrastructure.- [COMMAND_EXECUTION]: Performs a mandatory smoke test by compiling the current project's source code into a temporary binary and executing it to verify successful startup before initiating the release.- [PROMPT_INJECTION]: Processes untrusted data from Git commit messages and diffs to generate user-facing changelog entries, creating a surface for indirect prompt injection attacks. - Ingestion points: Git history and code diffs accessed via
git logandgit diff(SKILL.md). - Boundary markers: None; commit content is directly interpolated into the changelog without delimiters or instructions to ignore embedded commands.
- Capability inventory: Significant capabilities including file system writes, binary execution, and the ability to push commits and create pull requests.
- Sanitization: No validation or sanitization is performed on commit messages before they are processed by the agent.
Audit Metadata