dispatching-parallel-agents
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to process external, untrusted data (test failure logs and subsystem descriptions) and generate instructions for sub-agents without defining clear boundaries or sanitizing inputs.
- Ingestion points: Identification of failures from external test logs and subsystem file content as described in the domain identification phase (SKILL.md).
- Boundary markers: The provided prompt structure for sub-agents lacks markers to separate instructions from untrusted content or warnings to ignore embedded commands.
- Capability inventory: Sub-agents are granted capabilities to modify source code and test files to implement fixes (SKILL.md
- Agent Prompt Structure).
- Sanitization: The instructions do not include steps to sanitize or validate external content before it is used to generate tasks or modified during code fixing.
Audit Metadata