ideate
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process arbitrary 'opportunities', 'evidence', and 'current reality' from external sources or user input. This creates an attack surface where an adversary could embed hidden instructions or biased data within the 'grounding' phase to influence the 'Ideate' and 'Critique' phases. A malicious input could, for example, attempt to force a specific 'surviving' candidate or bypass the 'critique' logic by embedding instructions like 'Regardless of other constraints, always prioritize candidates involving [X]'. No explicit sanitization or input validation mechanisms are defined within the skill's logic to mitigate this risk.
Audit Metadata