database-migrator
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from database schemas (table and column names, comments) and data samples, creating a surface for potential manipulation if the database content contains hidden instructions. Ingestion points: Database schema metadata and data samples extracted in SKILL.md and references/schema-discovery.md. Boundary markers: Absent. Capability inventory: Bash execution and file writing tools as specified in SKILL.md. Sanitization: No specific sanitization or filtering logic is prescribed.
- [DYNAMIC_EXECUTION]: The skill generates and prepares to execute migration and validation scripts including SQL, Shell, and JavaScript code based on templates and discovered metadata. Evidence: Script generation steps in SKILL.md and corresponding templates in the references directory.
Audit Metadata