prisma-7
Pass
Audited by Gen Agent Trust Hub on Aug 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: All content was verified as technical documentation and implementation tutorials. No malicious scripts or executable code were found within the skill files.
- [SAFE]: Hardcoded credentials flagged by automated scanners were confirmed to be benign placeholders (e.g., 'johndoe:mypassword') used for educational purposes in the documentation.
- [SAFE]: The skill incorporates robust safety guardrails for AI agents, specifically instructing them to pause and request explicit user confirmation before running destructive database commands.
- [SAFE]: External dependencies and URLs point to trusted technology providers (Prisma, PostgreSQL, MongoDB, Microsoft) and official software repositories.
- [SAFE]: The instructions promote secure development practices, including the use of environment variables for secrets and the implementation of input sanitization to mitigate injection risks.
Audit Metadata