drizzle
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains extensive documentation and examples for Drizzle ORM. It promotes secure development practices, such as the use of parameterized queries and the implementation of authentication for database proxies. No malicious patterns or dangerous hidden behaviors were detected.
- [CREDENTIALS_UNSAFE]: Deterministic detectors flagged database connection strings in multiple files, including
samples/read-replicas.mdandscripts/config.md. Analysis confirms these are generic documentation placeholders (e.g.,user:password) and not active credentials. The skill consistently instructs the user to use environment variables (process.env.DATABASE_URL) for actual connections. - [INDIRECT_PROMPT_INJECTION]: The skill serves as a reference for the agent to generate database code. (1) Ingestion points: Untrusted user prompts requesting Drizzle ORM implementation help. (2) Boundary markers: Not explicitly defined within the skill's reference content. (3) Capability inventory: The skill describes CLI command execution via
drizzle-kitand database operations through the Drizzle ORM API. (4) Sanitization: The documentation explicitly highlights Drizzle's automatic parameterization as a primary defense against SQL injection.
Audit Metadata