avaritiachaos/dsh-plugin-multimodal-bridge ↗★ 0
@shion-lab/dsh-plugin-multimodal-bridge
Dynamic multimodal-to-text projection and cross-model vision bridge for DeepSeek Harness (dsh), allowing seamless hot-switching between vision models (Gemini/Claude) and text-only models (DeepSeek).
AI Analysis
核心用途是解决纯文本模型无法处理历史会话中图片的问题。适合需要在 Gemini/Claude 等视觉模型与 DeepSeek 等纯文本模型间频繁切换会话的用户。
Install
This plugin has no verified bundle, or compatibility checks failed. Read the repository notes first. Read the full README ↗
README
Read the full README ↗配置项 (Cordis Schema)
| 配置项 | 类型 | 默认值 | 描述 |
|---|---|---|---|
placeholderTemplate | string | [图片内容：历史多模态图像...] | 降级为纯文本时的占位说明文本 |
preserveCaptions | boolean | true | 若图片带有已提取的 Caption/OCR 描述，优先保留真实文字内容 |
bypassAdmissionLock | boolean | true | 自动解除 DSH 官方对于纯文本模型选取的死锁报错 |
customVisionModels | string[] | [] | 自定义额外的 Vision 模型关键字匹配列表 |