replicate
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFENO_CODE
Full Analysis
- Documentation Only (SAFE): The skill consists entirely of Markdown documentation and does not include any executable scripts or configuration files.
- External References (SAFE): All links point to official Replicate domains (replicate.com, api.replicate.com) for API documentation and schemas.
- Indirect Prompt Injection (LOW): The skill defines workflows for fetching external model schemas and metadata from the Replicate API.
- Ingestion points: API responses from
api.replicate.com(model schemas and search results). - Boundary markers: None specified in the documentation.
- Capability inventory: The skill itself has no code and performs no actions; it relies on the agent's native ability to make HTTP requests.
- Sanitization: Not applicable as no code is provided, but the agent should treat API-returned schemas as untrusted data.
Audit Metadata