generate-release-notes
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from GitHub commit messages and pull request descriptions, creating a surface for indirect prompt injection.
- Ingestion points: Untrusted data enters the agent context via GitHub API calls (
gh api,gh pr view,gh pr list) in steps 3 and 5 of the workflow. - Boundary markers: The skill lacks explicit delimiters or instructions to treat the ingested commit/PR text as untrusted data, which could lead the agent to follow instructions embedded in that text.
- Capability inventory: The skill possesses file system write capabilities, including directory creation (
mkdir -p) and writing markdown files to thereleases-docs/directory. - Sanitization: No formal sanitization or filtering of external content is specified beyond manual review for typos.
Audit Metadata