agent-queen-coordinator
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- Prompt Injection (LOW): The instructions 'Establish dominance hierarchy', 'write sovereign status', and 'Bypass consensus' in 'Emergency Mode' are role-play patterns that attempt to override standard agent constraints and collective decision-making protocols within the swarm context.
- Indirect Prompt Injection (LOW): The agent functions as an orchestrator that ingests data from 'scouts' and 'workers' to issue shared 'Royal Directives'. 1. Ingestion points: Data provided by external agent roles such as scouts and explorers. 2. Boundary markers: None; directives are formed directly from interpreted intelligence without delimiters. 3. Capability inventory: Persistent memory storage and shared directive issuance via the 'mcp__claude-flow__memory_usage' tool. 4. Sanitization: No validation or escaping of 'subject' data is described before storage in shared coordination namespaces.
Audit Metadata