orchestrator
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements an orchestration loop that ingests and validates results from external subagents, creating a surface for indirect prompt injection.
- Ingestion points: Subagent research and execution results are handed back to the main thread for validation as described in
SKILL.md. - Boundary markers: The instructions do not define clear delimiters or boundary markers to distinguish between the orchestrator's system instructions and the untrusted data returned by subagents.
- Capability inventory: The skill possesses planning and validation capabilities, and delegates 'Research' and 'Act' phases to subagents.
- Sanitization: There is no evidence of sanitization, filtering, or instruction-aware parsing for the data returned by subagents before it is processed by the main orchestrator thread.
Audit Metadata