db-architect
Pass
Audited by Gen Agent Trust Hub on Aug 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill operates by providing structured instructions to the agent to generate static SQL and documentation files based on user requests. It does not perform network operations, download external scripts, or use obfuscation techniques. The instructions for writing the output to the local file system (e.g.,
docs/SCHEMA.md) are consistent with its primary purpose as a developer productivity tool. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied app ideas or PRDs to generate database schemas. This ingestion of external data represents a potential attack surface for indirect prompt injection.
- Ingestion points: External text input (app ideas/PRDs) processed at runtime (SKILL.md).
- Boundary markers: None; the skill interpolates user requests directly into the design task.
- Capability inventory: File system write access to create markdown documents in the project workspace (SKILL.md).
- Sanitization: None.
Audit Metadata