conversion-diagnostic-kit
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill framework describes ingesting external data sources like 'Voice of Customer' (VOC) and session replays, creating a surface where malicious instructions embedded in external feedback could influence the agent's reasoning. 1. Ingestion points: SKILL.md Signal Stack (analytics, heatmaps, session replay, VOC). 2. Boundary markers: Absent. 3. Capability inventory: Mentions the 'diagnose-conversion-drop' command. 4. Sanitization: Absent.
Audit Metadata