orchestration-workflow
Pass
Audited by Gen Agent Trust Hub on Apr 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill outlines a structured process for performance profiling and kernel design using sub-agents. It uses local file paths and internal tool calls for legitimate development tasks.- [PROMPT_INJECTION]: The orchestration workflow consumes data from developer logs and sub-agent outputs, representing a theoretical surface for indirect prompt injection. However, this is inherent to its functional purpose as an orchestrator.
- Ingestion points: Reads developer logs (docs/kernels/*.md) and communication contracts from sub-agents.
- Boundary markers: No specific delimiters are defined in the templates for these inputs.
- Capability inventory: File read/write operations for documentation and code generation; internal tool execution (/profile-kernel, /design-kernel).
- Sanitization: No explicit sanitization or filtering of external content is specified.
Audit Metadata