venice-decisions
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of untrusted user data via the 'state' parameter within an AI decision-making framework, which is a potential vector for indirect prompt injection.
- Ingestion points: The 'state' field in POST /api/v1/decisions requests as defined in SKILL.md.
- Boundary markers: The documentation specifies that questions are processed in isolation from each other, providing a basic level of boundary separation.
- Capability inventory: The skill utilizes network capabilities through standard tools (curl, fetch, requests) to communicate with the vendor's API (api.venice.ai).
- Sanitization: No specific input sanitization is demonstrated in the provided code snippets.
Audit Metadata