xsoc1/dsh-image-vision

The flagship DeepSeek and GLM chat models are text-only and cannot read images. ModLens is a plug-in vision engine that gives a text-only model sight. ModLens reads images pasted straight into the chat, no saving to a file and passing a path first.

npm i -g @deepseek-ai/dsh dsh plugin --profile web add "github:xiaoshihou514/dsh-weixin#main"

DeepSeek Harness（DSH）任务状态通知插件 / Task-status notification plugin for DeepSeek Harness (DSH) 会话需要交互、任务完成、任务出错时，通过系统通知（OS toast）、浏览器通知、飞书机器人（webhook）推送提醒——不再依赖盯着会话列表的圆点。 Get notified via system notifications, browser notifications, and a Feishu (Lark) bot whenever a session needs attention, a task completes, or a task errors.

The main model used by DeepSeek Harness does not always support images. When a conversation involves a screenshot, photo, chart, or interface, dsh-open-eyes can send the image to a separately configured multimodal model and return its analysis as text to the current conversation.

The adapter composes the harness in-process and maps its session-event log onto the full ACP vocabulary: streamed text and reasoning, tool calls with diffs and display terminals, plans, permission requests, session modes, config options, slash commands, skills, and MCP servers. Credentials never touch your editor config — it reuses the key you saved in the dsh Web UI, or dsh-acp login saves one to the same store.

Free vision & image generation for DeepSeek Harness — paste an image into any chat, even text-only sessions. GLM-4V-Flash / Qwen3-VL / Gemini failover chain, ModLens-style structured evidence, Kolors generation. 免费读图·生图 · 三引擎容错 · 无 Key 入库