summarize-releases
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local commands such as
git,node,pnpm, andvaleto gather release evidence and validate documentation. \n- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from git history and changelogs which are ingested into the agent's context. \n - Ingestion points: Commit messages, changelog notes, and tags are read by
scripts/collect-releases.js. \n - Boundary markers: The data is presented in a Markdown report but lacks strict boundary delimiters to protect against embedded instructions. \n
- Capability inventory: The agent can run shell commands, write files to the repository, and call other local skills. \n
- Sanitization: No sanitization is performed on the data retrieved from the git repository.
Audit Metadata