Skills
Navigate
skills/veniceai/skills/venice-image-edit/Gen Agent Trust Hub

venice-image-edit

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill defines parameters for image editing and composition that ingest potentially untrusted data.
  • Ingestion points: The prompt and image (specifically https:// URL variants) fields in the API requests for /image/edit, /image/multi-edit, and /image/background-remove as described in SKILL.md.
  • Boundary markers: The skill uses structured JSON to delimit fields, but the content of the prompt and images is parsed by an AI model.
  • Capability inventory: The skill uses network capabilities (curl) to interact with the external api.venice.ai service.
  • Sanitization: The documentation highlights a safe_mode parameter (defaulting to true) provided by the platform to filter adult content, though this does not strictly prevent adversarial prompt instructions.
  • [COMMAND_EXECUTION]: The documentation includes multiple curl command examples (e.g., in the /image/edit and /image/upscale sections) to demonstrate API integration. These examples are standard for developer documentation and use environment variable placeholders ($VENICE_API_KEY) rather than hardcoded credentials.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 09:34 PM
Security Audit — agent-trust-hub — venice-image-edit