qrspi-iterate-research
Pass
Audited by Gen Agent Trust Hub on Jun 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns or security risks identified. The skill is designed for updating codebase research documents and stays within the local project context.
- [PROMPT_INJECTION]: The skill processes user feedback to guide research updates, which is an indirect prompt injection surface. 1. Ingestion points: User feedback provided to the agent as described in the process. 2. Boundary markers: Absent. 3. Capability inventory: File reading/editing and spawning sub-agents (codebase-locator, codebase-analyzer, codebase-pattern-finder). 4. Sanitization: Absent, though the skill requires file:line references for every new claim, providing a verification mechanism.
Audit Metadata