A DSH plugin for local ComfyUI image generation, inline results and optional tiled upscaling.
DSH integration
Native runtime
Author-claimed
Safety audit
Unaudited
Last verified
2026-10-18
License
MIT
01What can it help you accomplish?
Generate images from text or reference images in DSH
Images saved to disk and displayed in the conversation
DSH users with a local ComfyUI setup and Qwen models
02How to install into DeepSeek Harness
Prerequisites
- A local ComfyUI instance at 127.0.0.1:8188 and the three Qwen model files listed in the README.
Installation steps
- 01
From the local repository copy, run `node scripts/install.mjs --profile web` to install into the web profile.
- 02
Fully quit DSH and reopen it after installation.
Verify the integration
- Run `node scripts/setup.mjs --check` to inspect missing setup components without changing them.
03DSH integration and capability boundaries
DSH image-generation plugin using local ComfyUI
Text-to-image and image-to-image
Use Qwen-Image 2.1 through local ComfyUI; save generated images and show them in chat.
Post-generation upscaling
Set `upscaleAfter` to run tiled upscaling after generation; an upscale failure leaves the original image saved.
04Who is it for? When not to use it?
Good for
- DSH users with a local ComfyUI setup and Qwen models
Not for
- Acceleration-package installation is enabled by default: accelAutoInstall=true may run pip install --upgrade or git clone. Set accelAutoInstall=false to disable this without disabling image generation.
- Remote ComfyUI/hipfire endpoints require explicit allowRemoteComfy opt-in; enabling it sends prompts and images to the chosen address. It is off by default.
05Compatibility, maintenance and safety notes
- Acceleration-package installation is enabled by default: accelAutoInstall=true may run pip install --upgrade or git clone. Set accelAutoInstall=false to disable this without disabling image generation.
- Remote ComfyUI/hipfire endpoints require explicit allowRemoteComfy opt-in; enabling it sends prompts and images to the chosen address. It is off by default.
README author claims; not independently tested
06Frequently asked questions
What does dsh-qwen-paint add to DeepSeek Harness?
Text-to-image and image-to-image generation through local ComfyUI and Qwen-Image 2.1, with images saved and shown in the conversation.
How do I install it into the DSH web profile?
Prepare the README-listed ComfyUI and model files, run `node scripts/install.mjs --profile web` from the repository copy, then fully quit and reopen DSH.
Does using it in DSH change the local environment?
Automatic acceleration-package installation is enabled by default and may run pip install --upgrade or git clone. Set accelAutoInstall=false to turn it off without disabling image generation.
07Related DSH workflows
modlens
by liustack
Vision bridge for text-only models: paste an image, get structured JSON evidence (OCR, layout, semantics).
agent-vision-toolkit
by anionex
Vision toolkit and skills that give text-only LLMs eyes — multi-image understanding, image Q&A, OCR, frontend UI restoration and GUI automation, with optional agent integration.
dsh-vision-router
by ysr666
Eyes for text-only DeepSeek Harness agents: built-in free vision chain (no key) + pixel-level vision tools (Q&A, grounding, crop, pixel diff, colors, OCR, SVG trace, cutout, screenshots). One-command install, no Python, image turns work like ordinary tool-calling turns.
dsh-vision-toolkit
by anionex
为纯文本 DSH Agent 提供 10 个结构化视觉工具：意图感知图片问答、长截图 OCR、原始像素 grounding、UI 还原、像素 diff 等
08Data and sources
面向 **DeepSeek Harness** 的本地图像生成插件。
**MIT** —— 见 [LICENSE](LICENSE)。
This page is generated from the project’s public documentation, repository metadata and a structured parse of DSH Plugins; last verified on 2026-10-18. Found an error? Submit a correction.
Best DeepSeek Harness Plugins
Twelve plugins worth installing first — picked from the whole catalog, across every category.