oma-db
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security risks identified. The skill focuses on providing expert database design and architecture guidance.
- [COMMAND_EXECUTION]: The skill instructs the agent to run local search commands (e.g.,
rg) and project-specific migration or query tools. This is a standard capability required for database management tasks and is performed within the user's project context. - [PROMPT_INJECTION]: Indirect prompt injection surface analysis:
- Ingestion points: Reads schemas, migrations, and query files from the codebase (SKILL.md, execution-protocol.md).
- Boundary markers: Employs structured output templates and a self-verification checklist to guide response generation (document-templates.md, checklist.md).
- Capability inventory: Reading and writing codebase files, executing shell commands for migrations and query planning (SKILL.md).
- Sanitization: Includes explicit guidance on preventing SQL injection in application code and maintaining data integrity (anti-patterns.md).
Audit Metadata