sota-api-design
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill content consists exclusively of markdown-based instructions, design rules, and audit checklists. No executable scripts or malicious patterns were identified. The rules promote industry-standard security practices such as HMAC signing, SSRF defense, and tenant isolation.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze untrusted code and specifications provided by the user in a project environment.
- Ingestion points: The skill instructions direct the agent to read route definitions, middleware, proto files, and schema definitions (e.g., rules/01, rules/03).
- Boundary markers: Instructions in SKILL.md explicitly warn the agent to "verify in code, don't trust docs or comments," which serves as a defensive measure against malicious instructions potentially embedded in data comments or documentation.
- Capability inventory: The skill's functionality is limited to static analysis, file reading, and logic verification within the provided project structure.
- Sanitization: The skill emphasizes validating the implementation against the contract (OpenAPI/SDL/Proto) to ensure consistency and detect unauthorized logic drift.
Audit Metadata