domain-modeling
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user input and existing project files (source code, ADRs, and glossaries) to maintain a domain model. This creates a surface where content from the codebase could influence the agent's output.
- Ingestion points: User conversation,
GLOSSARY.md,GLOSSARY-MAP.md, source code within thesrc/directory, and architectural records indocs/adr/. - Boundary markers: None defined. The skill assumes trust in the project's source files and user input.
- Capability inventory: Reading and writing Markdown files (
GLOSSARY.md,ADRfiles) and reading repository source code. - Sanitization: No explicit sanitization or filtering of ingested data is described; the agent is instructed to focus on domain terminology and avoid implementation details.
Audit Metadata