planetscale-schema-recommendations-agent-loop
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements secure operational practices by enforcing human-in-the-loop verification for all database schema modifications. It explicitly forbids the autonomous application of production changes, directing the agent to instead create pull requests, issues, or development branches for review.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input from external database recommendations which could theoretically contain malicious instructions. However, the risk is mitigated by procedural instructions that require all output to be reviewed as code changes rather than executed directly.
- Ingestion points: External schema recommendations, DDL suggestions, and query pattern evidence (SKILL.md).
- Boundary markers: The skill does not define specific syntax boundaries for the ingested data.
- Capability inventory: The skill is restricted to codebase analysis and the creation of pull requests or issues.
- Sanitization: No explicit data sanitization is mentioned, relying instead on the mandatory human approval process.
Audit Metadata