koa-typescript
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides templates for building web APIs that ingest external data, which is a standard attack surface for web applications.
- Ingestion points: Middleware and route examples in
SKILL.mdprocessctx.request.bodyandctx.headers.authorization. - Capability inventory: The skill outlines web server functionality including routing, request handling, and middleware execution.
- Sanitization: The instructions explicitly recommend using
zodfor schema validation andkoa-helmetfor security headers, which are industry-standard mitigations. - Boundary markers: The skill follows standard HTTP request/response boundary patterns.
Audit Metadata