onboarding-teardown
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data including user descriptions and complex Flow Builder JSON configurations, which represents an attack surface for indirect prompt injection.
- Ingestion points: The skill accepts user descriptions, application screens, and Flow Builder JSON files to drive its analysis sequence.
- Boundary markers: No explicit delimiters or instructions are provided to the agent to isolate user data from core logic or to ignore instructions embedded in the analyzed content.
- Capability inventory: The skill generates prioritized hypotheses and can invoke the flow-generator skill to build or modify application flows based on the analysis.
- Sanitization: The instructions do not define validation or sanitization steps for the ingested JSON data or text before processing.
Audit Metadata