async
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill coordinates the delegation of work to subagents based on user input. While it defines a clear structure for framing tasks, it does not explicitly instruct the agent to use protective delimiters or sanitization when passing user instructions to the subagent environment, creating a potential surface for indirect injection.
- Ingestion points: Task objectives and deliverables provided by the user (as described in SKILL.md and README.md).
- Boundary markers: None specified in the instructions for subagent invocation.
- Capability inventory: Spawning and launching subagents with scoped access (SKILL.md Operating Rule 3).
- Sanitization: No explicit sanitization or validation of delegated task content is mentioned.
Audit Metadata