release
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes various shell commands for repository management, including git operations (status, branch, commit, push), GitHub CLI commands (pr create, pr merge, release create), and npm lifecycle scripts (build, test, version). These are standard operations for a software release process.
- [EXTERNAL_DOWNLOADS]: The workflow involves downloading packages from the official npm registry using
npx(e.g.,npx vitest,npx agentic-qe) and installing dependencies in temporary directories for verification. These operations target well-known, trusted registries. - [DYNAMIC_EXECUTION]: The skill executes locally built JavaScript artifacts (e.g.,
node dist/cli/bundle.js) to verify the build before release. It also usesnode -pto evaluate short JavaScript snippets for reading the version frompackage.json. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from local repository files, including
package.json,CHANGELOG.md, and.claude/skills/skills-manifest.json. While these files could theoretically contain malicious instructions, the skill implements multiple 'STOP' points for mandatory user review and confirmation before any critical actions (like merging code or publishing) are taken.
Audit Metadata