architect
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes existing codebase subsystems during its initial grounding and sketching phases to synthesize new architectural designs.
- Ingestion points: Phase A (Grounding) reads subsystem data, and Phase B (Sketch) passes these grounding artifacts to secondary agents for design generation.
- Boundary markers: The instructions do not specify any delimiters or safety warnings to ensure the agent ignores instructions potentially embedded within the analyzed code or documentation.
- Capability inventory: The skill is capable of generating module maps, type sketches, and implementation code (Phase D) that are written back to the codebase.
- Sanitization: There is no evidence of filtering or sanitization performed on the content retrieved from the grounding artifacts before it influences the architectural output.
Audit Metadata