multi-agent-coordination
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines workflows for ingesting and comparing outputs from sub-agents, which represents a potential indirect prompt injection surface.
- Ingestion points: The agent is instructed to collect conclusions from independent sub-agents and merge their findings (Sections 3 and 5).
- Boundary markers: The skill implements a '6-Section Contract' for delegation prompts and mandates specific output sections (Summary, Files, Verification, Risks) to structure external data.
- Capability inventory: The coordination logic facilitates the use of tools such as Bash, Grep, and Read through delegated sub-agents.
- Sanitization: The skill mitigates risks by requiring an 'Output Grading' system (Section 1) where results are verified against file references and command outputs rather than being accepted as raw text.
Audit Metadata