definition-of-done
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze untrusted data from external sources during the auditing process.
- Ingestion points: External data is processed from repository artifacts including pull request review JSON (via
gh pr view), git diff outputs, and coverage report files (e.g.,coverage/lcov.info). - Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when the agent processes the content of PR reviews or commit diffs.
- Capability inventory: The skill encourages the use of CLI tools like
ghandgitand direct file system reads to verify audit checkpoints. - Sanitization: There are no explicit instructions for sanitizing or validating the content of the external artifacts before they are interpreted by the agent.
- [COMMAND_EXECUTION]: The skill guidelines recommend using specific CLI commands to gather evidence for audits.
- Evidence:
SKILL.mddescribes usinggh pr view --json reviews,reviewDecision,gh pr view --json statusCheckRollup, andgit diff --name-only <base>..<head>to retrieve repository state.
Audit Metadata