specialized-perfect-agent-orchestrator
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest high-level user objectives and decompose them into specialized tasks for sub-agents. This workflow presents an inherent risk of indirect prompt injection where adversarial input could influence the behavior of downstream agents.
- Ingestion points: User objectives are accepted as input via the
run.shscript (through stdin or command-line arguments) and passed to the skill runner. - Boundary markers: The skill instructions do not specify any delimiters or safety warnings to distinguish between system instructions and user-provided task data.
- Capability inventory: The orchestrator has the capability to spawn sub-agents via the
SubAgentMiddleware, use persistent filesystem storage, and interface with external MCP servers. - Sanitization: There are no visible mechanisms for sanitizing, escaping, or validating user input before it is interpolated into the orchestration workflow.
Audit Metadata