andes-release

Pass

Audited by Gen Agent Trust Hub on Sep 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides bash script blocks intended for repository inspection, tag backfilling, and workflow deployment.
  • These scripts leverage standard tools such as git, jq, and the GitHub CLI (gh) to perform legitimate release management tasks.
  • [EXTERNAL_DOWNLOADS]: The GitHub Actions workflow template included in the skill references actions/checkout@v6.
  • This is a standard, official action from GitHub used to provide the runner access to the repository source code during the release process.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external data from the repository to automate release documentation.
  • Ingestion points: The skill reads project metadata from apps/mobile/app.json and uses merged Pull Request titles to generate release notes.
  • Boundary markers: The skill does not explicitly define custom boundary markers for the ingested data.
  • Capability inventory: The skill has the capability to create git tags, push to remote repositories, and publish GitHub Releases.
  • Sanitization: The skill relies on the built-in --generate-notes functionality of the GitHub CLI, which provides standard formatting for release descriptions based on commit history.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 5, 2026, 07:06 PM
Security Audit — agent-trust-hub — andes-release