domain-modeling
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from the repository (glossaries, ADRs, and source code) to synchronize domain terminology, creating an attack surface for indirect prompt injection.
- Ingestion points: Reads various project files from the repository to extract and verify domain concepts.
- Boundary markers: No specific delimiters or safety instructions are defined for the agent when ingesting these files.
- Capability inventory: The skill possesses file read/write capabilities and can interact with external issue trackers with user permission.
- Sanitization: While no specific sanitization is applied to input data, the skill explicitly warns against shell injection when generating output for external platforms.
Audit Metadata