v1.0.0

🎉 v1.0.0 — First stable release

On-demand vision for text-only DeepSeek Harness (DSH) sessions: images become text markers in the model input, and a vision_describe tool sends only image + question to any OpenAI-compatible vision endpoint — no long context ever reaches the vision model.

✨ What's new

• Default vision model gpt-5.6-terra (schema default + settings UI placeholder updated)
• Docs & discoverability
  • Badge row in README (language / npm version / downloads / GitHub stars / license)
  • ❤️ Sponsors section — thanks to xiaoyaoapi for their API donation
  • Search-keywords section (GitHub indexing) + expanded npm keywords
  • Repo description, homepage and 16 topics set on GitHub

📦 Install

dsh plugin --profile web add @dsh-extension/dsh-vision-bridge

or

npx @deepseek-ai/dsh@0.1.0-rc.6 plugin --profile web add @dsh-extension/dsh-vision-bridge

🔗 Links

• npm: https://www.npmjs.com/package/@dsh-extension/dsh-vision-bridge
• 中文文档: https://github.com/sfyyy/dsh-vision-bridge/blob/main/README.zh-CN.md