db-migrate

Pass

Audited by Gen Agent Trust Hub on Nov 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns detected. The skill instructions focus on standard DevOps procedures for database management and incorporate extensive safety guardrails.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a potential attack surface for indirect prompt injection, as it processes local repository files to determine migration patterns.
  • Ingestion points: The skill reads repository-specific documents such as CLAUDE.md, backend-patterns-reference.md, and configuration files like docs/project-config.json (SKILL.md).
  • Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded prompts within the ingested project documentation.
  • Capability inventory: The skill utilizes shell execution (via migration tooling) and file system writes to create and apply database changes.
  • Sanitization: No specific content sanitization or validation logic is defined for the reference documents. However, the risk is effectively mitigated by the skill's requirement for a specialized database-admin sub-agent review and explicit user approval before any non-local migration is applied.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 5, 2026, 06:42 PM