definition-of-ready-done
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input—such as team names, delivery models, and descriptions of current process pain points—and interpolates this data into a final report template. The absence of explicit boundary markers or sanitization techniques creates a surface for indirect prompt injection.
- Ingestion points: User input collected during Phase 1 (Setup) and Phase 2 (DoR checklist adaptations).
- Boundary markers: The instructions do not define specific delimiters or "ignore instructions" warnings to isolate user-provided data from the agent's logic.
- Capability inventory: The skill is restricted to generating text reports and rendering Mermaid diagrams. It does not possess capabilities for network exfiltration, arbitrary command execution, or sensitive file system access.
- Sanitization: There is no evidence of input validation, filtering, or escaping for the external content provided by the user.
Audit Metadata