prepare-release
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes multiple shell commands to interact with the local Git repository, including
git describeto find tags,git logto extract commit history, and a sequence ofgit checkout,git add,git commit, andgit pushto manage release branches. These operations are essential for the skill's stated purpose of automating releases. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the processing of untrusted data.
- Ingestion points: The agent reads commit titles and bodies (specifically content within
release-noteblocks) viagit login SKILL.md Step 2. - Boundary markers: There are no explicit instructions or delimiters provided to the model to ignore potential commands embedded within the commit messages during the summarization and translation phase (Step 3).
- Capability inventory: The agent has the ability to write to the local filesystem (
package.json,electron-builder.yml) and push changes to a remote repository via Git commands. - Sanitization: No sanitization or validation of the commit message content is performed before it is processed by the LLM for release note generation.
Audit Metadata