Overview
dsh-image-auto-describe
Image auto-describe provider for the DeepSeek Harness apiproxy admission seam: transcribes pasted images through configurable vision routes (Qwen first, GLM fallback) so text-only session models still receive image prompts.
README / EN
Package documentation
Registry summary
Image auto-describe provider for the DeepSeek Harness apiproxy admission seam: transcribes pasted images through configurable vision routes (Qwen first, GLM fallback) so text-only session models still receive image prompts.
dsh.pub verifies the pinned bundle contract, runtime facts, and distribution semantics. The complete README remains in the source repository.
Read the full README on GitHubLIMITATIONS
Known limitations
- 转写文本替代了模型可见消息中的原图；视觉模型未转写出的图片细节，会话模型无法回答（对话中仍可预览原图）。 - 未配置路由或全部路由失败（无密钥、无额度）时，消息以原错误拒绝，而非降级放行。 - 路由必须已存在于组合好的 llm 能力中（由部署的 llm 提供者插件注册）；本插件不注册提供者。