find-models
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to fetch and process external model metadata, including descriptions, tags, and schemas, which could be crafted to influence agent behavior.
- Ingestion points: Replicate API endpoints including
GET /v1/search,GET /v1/collections, andGET /v1/modelsmentioned inSKILL.md. - Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands in the retrieved data.
- Capability inventory: The skill prepares the agent to select and configure models for execution based on the retrieved external data.
- Sanitization: There is no evidence of metadata sanitization or validation of the retrieved model schemas before they are processed by the agent.
Audit Metadata