prepare-release
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes multiple shell commands to manage the Git repository and interface with the GitHub API.
- Uses
git fetch,git log,git checkout,git commit, andgit pushto manage release branches and metadata. - Uses
gh release list,gh repo view, andgh apito interact with GitHub platform services. - Executes local scripts using Node.js (
scripts/release/sync-release-history.js,scripts/release/validate-prepared-release.js, etc.) and runs build commands viapnpm. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the Git history to generate bilingual release notes.
- Ingestion points: Commit messages and content within
```release-notecode blocks are collected in Step 2 and passed to the LLM. - Boundary markers: The instructions do not specify any delimiters or safety warnings to distinguish between instructional prompts and the collected commit data.
- Capability inventory: The skill possesses the ability to modify project configuration files (
electron-builder.yml,package.json), create release branches, and push changes to the remote repository. - Sanitization: No sanitization, escaping, or filtering of the extracted commit content is performed before it is processed by the agent.
Audit Metadata