Sn Image Base

This is infrastructure, not a user-facing tool. It's the tier-0 foundation for the SenseNova-Skills project, exposing three raw API wrappers: text-to-image generation, VLM-based image recognition, and LLM text optimization. No preprocessing, no guardrails, just direct calls to backend services. You'd use this if you're building higher-level skills that need clean separation between orchestration logic and API plumbing. The parameter design is thoughtful, with three-level fallbacks for auth and endpoints, plus support for both OpenAI and Anthropic protocol styles. It's deliberately bare-bones, which is exactly what a base layer should be, but that also means you'll need to wrap it with something smarter before exposing it to end users.

Install to Claude Code

npx -y skills add opensensenova/sensenova-skills --skill sn-image-base --agent claude-code