flutter-parallel-agents
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a prompt template for sub-agents that involves direct interpolation of external data.
- Ingestion points: The section 'Agent Prompt Structure' in
SKILL.mdinstructs the agent to 'Paste specific requirements from plan' into the sub-agent's context. - Boundary markers: While markdown headers are used to structure the prompt, there are no specific boundary markers or instructions telling the sub-agent to disregard instructions contained within the requirements data.
- Capability inventory: Sub-agents created using this pattern have the capability to create/modify files and execute shell commands (e.g.,
flutter analyze). - Sanitization: There is no mention of sanitizing, escaping, or validating the input requirements before they are passed to the sub-agents.
Audit Metadata