enhance-orchestrator
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill coordinates sub-agents to analyze project files (Phase 2) and can automatically apply 'HIGH certainty' fixes based on their findings (Phase 8). This creates a pipeline where malicious content inside analyzed files could attempt to influence the agent's behavior.
- Ingestion points: Project files discovered via
Globpatterns in Phase 2 (plugins, agents, docs, prompts, etc.). - Boundary markers: The
Task()prompts used to spawn sub-enhancers do not include explicit delimiters or instructions to ignore embedded commands within the analyzed content. - Capability inventory: The skill has the capability to delegate file modifications to sub-agents via
Task()calls in Phase 8 if the--applyflag is used. - Sanitization: There is no evidence of sanitization or escaping of the aggregated findings before they are passed back to sub-agents for fixing.
Audit Metadata