address-github-comments
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external GitHub comments and review threads, which could contain instructions designed to influence the agent's behavior.
- Ingestion points: The skill fetches external content via
gh pr view --commentsand GraphQL queries forpullRequest.reviewThreadsin the 'Inspect Comments' workflow step. - Boundary markers: There are no explicit instructions or delimiters defined to separate the instructions within the comments from the agent's primary system instructions.
- Capability inventory: The skill possesses file-write capabilities (Step 3: 'Apply Fixes') and network-write capabilities (Step 4: 'Respond to Comments' via
gh pr comment). - Sanitization: The workflow does not include sanitization or validation steps to filter potential prompt injection attempts embedded within the retrieved GitHub comments.
Audit Metadata