contributions-hub
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-generated content from external sources (GitHub discussions, issues, and pull requests) to generate summaries and draft welcoming responses.
- Ingestion points: Untrusted data is ingested when scanning GitHub Discussions (Workflow Modes A, C, D, H) and Contributor Activity/PRs (Workflow Modes F, G).
- Boundary markers: The instructions do not define explicit boundary markers, delimiters, or 'ignore' instructions for the agent when processing this untrusted external data.
- Capability inventory: The skill possesses the capability to write to the repository environment by creating issues and posting comments to discussions and pull requests.
- Sanitization: No specific requirements for sanitization or escaping of the fetched external text are provided before the content is interpolated into generated summaries or drafts.
Audit Metadata