Overview
@kbpoyo/dsh-image-bridge
DSH Web plugin that lets text-only models see images: paste images in the chat and send without file paths; the model discovers its own vision tools. Multimodal models pass through natively.
README / EN
Package documentation
Registry summary
DSH Web plugin that lets text-only models see images: paste images in the chat and send without file paths; the model discovers its own vision tools. Multimodal models pass through natively.
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
- **自定义供应商模型的能力上报**：凡经 settings 配置的自定义供应商（provider），其模型声明未包含 `input: ['text','image']`（或适配器未上报 image 能力）时，适配器一律将其上报为 `inputModalities=['text']`。此类模型下桥接按钮会**如实显示**（插件忠实读取适配器上报，无法判定其为视觉模型）——这是配置层事实，与插件无关。要让这类模型正确隐藏按钮，需在模型声明中补 `input` 字段，或改用适配器托管目录、正确上报能力的模型。 - **原生门禁不绕过**：DSH 原生"会话日志含原生图片后禁止切回纯文本模型"的限制保持不变（切回需 `/compact` 或新会话）。 - **subagent 传图不在覆盖范围**：父代理向子代理发送的含图消息不经 `session.prompt`，仍走原生门禁。