orchestration-workflow
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The orchestration workflow involves reading developer logs and sub-agent reports to generate instructions for subsequent tasks. This content is used to construct prompts for the kernel-designer agent.\n
- Ingestion points: Reads developer logs from
docs/kernels/and analysis reports from the profiler sub-agent.\n - Boundary markers: No specific delimiters are defined in the launch templates to separate ingested data from instructions.\n
- Capability inventory: The skill coordinates launching sub-agents and writing updates to project documentation and knowledge base files.\n
- Sanitization: No explicit sanitization or validation of ingested log content or agent reports is specified in the workflow instructions.
Audit Metadata