kysely-schema
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the ingestion of external data.
- Ingestion points: According to
SKILL.md, the agent processes external inputs including "MySQL DDL", "CREATE TABLE statements", and "table structure screenshots". - Boundary markers: The instructions do not include specific boundary markers or warnings to disregard potential instructions hidden within the input SQL comments or metadata.
- Capability inventory: The skill is authorized to modify sensitive project files, including
src/types/database.d.tsand various service files (*-modify.ts,*-query.ts). An injection attack could potentially trick the agent into writing malicious logic into these service files. - Sanitization: There are no documented steps for sanitizing or escaping the content of the DDL statements before they are used to generate code.
Audit Metadata