skills/yogsoth-ai/de-anthropocentric-research-engine/coverage-white-space-search/Gen Agent Trust Hub
coverage-white-space-search
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data (reference_items and performance_records) without explicit sanitization or boundary markers.
- Ingestion points: SKILL.md (via input contract fields: reference_items, performance_records).
- Boundary markers: Not present; instructions do not define delimiters for processed data.
- Capability inventory: None; the skill logic is restricted to analysis and synthesis without network or file system write access.
- Sanitization: Not present; the skill accepts inputs as-is for mapping.
Audit Metadata