github-codebase-briefing
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves and analyzes data from external, attacker-controllable sources (GitHub issues, pull requests, and code diffs) which could contain malicious instructions meant to subvert the agent's logic or bias its report.
- Ingestion points: Content is ingested via
gh issue list,gh pr list, andgh pr diffcommands as described inSKILL.md. - Boundary markers: The instructions do not define any delimiters or provide guidance for the agent to distinguish between its processing logic and the data from the repository.
- Capability inventory: The agent uses the
ghCLI to read repository metadata, structure, and content. - Sanitization: No sanitization or filtering of external issue/PR text is performed before the agent performs its 'Deep Analysis Logic'.
Audit Metadata