Vision & Input What it does modlens Gives text-only DeepSeek models plug-in vision: paste a screenshot, get structured evidence. Curated collection
DeepSeek Harness plugins for design
A curated collection of dsh plugins for design work: vision bridges that read screenshots, canvases and generative UI the agent can draw on, design review tools, and workbenches that preview it all. DeepSeek Harness reads the same SKILL.md format as Claude Code and Codex, so your design skill library carries over.
Vision & Input What it does modlens Gives text-only DeepSeek models plug-in vision: paste a screenshot, get structured evidence. 
Vision & Input What it does dsh-vision-toolkit Ten vision tools for UI restoration, grounding and pixel-diff verification. 
Vision & Input What it does dsh-ui-spec Turns UI screenshots into implementation-grade specs: tokens, spacing scale, layout grid. 
Vision & Input What it does dsh-media-skills Free eyes and a free paintbrush: paste-image reading plus watermark-free image generation. 
Vision & Input What it does dsh-vision-router Free, keyless eyes for text-only agents: a built-in vision chain plus eleven pixel-level tools. 
Canvas & Generative UI What it does dsh-openpencil The agent designs on a real, editable vector canvas instead of returning static images. 
Canvas & Generative UI What it does dsh-visualize The model draws interactive HTML cards directly into the conversation stream. 
Canvas & Generative UI What it does dsh-genui Thirty-plus interactive components rendered inline in replies, with an action loop back to the model. 
Canvas & Generative UI What it does dsh-openmaic Slides, interactive widgets and full playable lessons, rendered from agent-written JSON. 
Canvas & Generative UI What it does dsh-diagram Turn any article in the session into an editable Excalidraw canvas, not a disposable image. 
Canvas & Generative UI What it does deepseek-harness-genui The task grows a focused React interface — and what you choose in it flows back to the agent. 
Design Workflow What it does dsh-web-review Point at elements on a live page, annotate visually, and the agent edits the source. 
Design Workflow What it does dsh-figma-to-lottie Compiles SVG paths and keyframes into self-contained Lottie animations from the conversation. 
Design Workflow What it does dsh-plugin-claude-bridge Your Claude Code skills, memory and global instructions, available in the harness with zero migration. 
Design Workflow What it does dsh-annotate Point at browser elements and send the agent structured facts, not vague descriptions. 
Design Workflow What it does dsh-hyperframes HyperFrames by HeyGen, ported: five official skills that turn HTML into finished video. 
Workspace & Preview What it does dsh-web-ui The ecosystem’s biggest UI kit: task board, preview panel, git graph and a skin center. 
Workspace & Preview What it does DSH-better-sidebar A full workbench in the sidebar: file explorer, rich previews, terminal, git and a browser. 
Workspace & Preview What it does dsh-web-preview A glass side panel that previews the workspace, runs the project and annotates elements. Next step
Skip the setup. Design with DeepSeek Harness inside Open Design
Open Design is the open-source, agent-native design workspace that runs around DeepSeek Harness. It keeps your systems, skills and templates consistent, so the agent ships work you own.