domain-modeling
Pass
Audited by Gen Agent Trust Hub on Aug 13, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input and project source code to generate and update domain model documentation and Architectural Decision Records (ADRs). While this represents an attack surface for indirect prompt injection, the risk is inherent to the skill's primary purpose of analyzing and documenting development projects.
- Ingestion points: User conversation and project source code files are read to extract domain concepts.
- Capability inventory: The skill utilizes file reading capabilities to scan code and file writing capabilities to update
CONTEXT.mdand create ADR files. - Boundary markers: The skill instructions do not explicitly define boundary markers for untrusted content, though they provide strict formatting templates (
CONTEXT-FORMAT.md,ADR-FORMAT.md) which serve as implicit structures. - Sanitization: No specific sanitization or escaping mechanisms are defined in the skill instructions; it relies on the underlying agent's standard text processing.
Audit Metadata