agent-swarm
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill identifies tools that ingest natural language tasks which could be manipulated if source data contains malicious instructions.\n
- Ingestion points: The
taskparameter in themcp__flow-nexus__task_orchestratetool (SKILL.md) accepts raw string input.\n - Boundary markers: No explicit delimiters or boundary markers are defined in the orchestration examples to separate user instructions from task data.\n
- Capability inventory: The skill possesses the capability to spawn new specialized agents via
mcp__flow-nexus__agent_spawnand coordinate multi-agent tasks.\n - Sanitization: The instructions do not specify any validation or sanitization routines for the data passed into the orchestration functions.
Audit Metadata