design-synthesis
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill processes untrusted external data which could contain malicious instructions to bias or subvert the design synthesis process.\n
- Ingestion points: Loads research data from
research_{slug}.mdand processes theRequestinput variable.\n - Boundary markers: Uses markdown headers and tables for structure, but lacks explicit delimiters or instructions for the agent to ignore embedded commands within the artifacts.\n
- Capability inventory: Employs
parallel-dispatchfor codebase scanning and performs file-write operations to createbrainstorm.md.\n - Sanitization: No input validation or sanitization logic is present for the ingested research content.\n- [Prompt Injection] (LOW): Use of authoritative command markers such as
[CRITICAL],[ALWAYS], and[NEVER]within the skill instructions, while used here for internal logic, mirrors patterns found in prompt injection attempts.
Audit Metadata