Setup & Installation
npx skills add https://github.com/veniceai/skills --skill venice-chat
https://github.com/veniceai/skills/tree/main/skills/venice-chat
View on GitHubWhat This Skill Does
Calls Venice's OpenAI-compatible chat completions endpoint with support for web search, end-to-end encryption, multimodal inputs, tool calling, structured output, and prompt caching. Venice-specific features like characters, X/Twitter search, and thinking controls are configured through a venice_parameters object or model name suffixes.
Covers the full Venice chat completions surface, including Venice-only parameters, multimodal inputs, and model suffixes, so you don't have to cross-reference multiple docs while building requests.
When to use it
- Sending multimodal prompts with images, audio, or video to Venice-hosted models
- Adding web search context to LLM responses with citation superscripts
- Getting structured JSON output from Venice models using json_schema format
- Applying Venice character personas to chat completions
- Caching large system prompts across repeated Venice API calls