pr-draft-summary
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from an existing codebase, which could contain malicious instructions designed to influence the agent's behavior.
- Ingestion points: The workflow (SKILL.md) requires inspecting repository instructions, files, and user changes.
- Boundary markers: The skill instructs the agent to preserve authorization boundaries and verify real output, though it lacks explicit delimiters for untrusted code content.
- Capability inventory: The skill has the ability to call
sandbase_runfor external capabilities and perform repository inspection. - Sanitization: There are no instructions for sanitizing or escaping the codebase content before processing it for summary generation.
Audit Metadata