release-nightly
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data that could contain malicious instructions intended to influence the agent's behavior during the release process.
- Ingestion points: Reads commit messages via
git log, project-specific release instructions inreleases/WRITING.md, and workflow files (e.g.,.github/workflows/nightly.yml). - Boundary markers: The skill does not define specific delimiters for separating these external inputs from its internal instructions.
- Capability inventory: The agent has the capability to execute
git commit,git push(including force-pushing the nightly tag), and GitHub CLI (gh) commands to manage releases and runs. - Sanitization: The skill includes a validation step to check that generated notes parse and do not contain forbidden content like HTML or images, but it does not explicitly mention escaping or sanitizing commit messages before processing.
Audit Metadata