curated collections / image-generation
DSH Image Generation Plugins: Chat-to-Image & Video Picks
Describe it in chat and watch it render. These plugins give DeepSeek Harness real image and video generation: ChatGPT-style chat-to-image, Vox-style explainer videos and local ComfyUI pipelines — ranked by which to install first.
Handing the agent a paintbrush is the hottest content-generation scene in the dsh ecosystem. The head pick, shanliuling/dsh-image-gen (★472 and climbing), makes generation feel like chatting with ChatGPT: produce, edit and compare images right in the conversation, with a 500+ prompt library and Seedream/Gemini/OpenAI routes. Moving from stills to motion? alisa0808/vox-director (★2,024) turns a single topic into a finished Vox-style paper-collage explainer video — storyboard, voiceover and captions included; dickpy/dsh-imagegen covers text-to-image and image-to-image in the Web GUI. Own a GPU? fandc520/dsh-comfyui (★83) bridges your local ComfyUI workflows as agent tools, and wbin0001/dsh-comfyui-canvas wraps them in a canvas IDE. One direction check: this page is about output — making the agent draw. Reading images (vision, OCR) is the input side, covered by our /learn/image-recognition guide; the video-generation walkthrough lands soon at /learn/video-generation.
In a hurry? Install shanliuling/dsh-image-gen — say what you want in chat and it renders, edits and compares, 500+ prompts bundled. For video, alisa0808/vox-director turns one topic into a full explainer. Local-first? fandc520/dsh-comfyui connects your own ComfyUI.
103 plugins3,089 stars
Synced from our GitHub list · Catalog updated 2026-09-25
Why it matters now
Chat-to-image is the lowest-friction start: shanliuling/dsh-image-gen keeps you inside the conversation — generate, refine, compare variants — while its 500+ prompt library does the prompt engineering for you; dickpy/dsh-imagegen adds point-and-click text-to-image and image-to-image in the Web GUI.
Stills are just the beginning: alisa0808/vox-director (★2,024) produces complete Vox-style paper-collage explainer videos from one topic, storyboard to captions, while Seedance- and MiniMax-based plugins such as dsh-short-video-studio and dsh-seedance2 bring video models into the same chat.
For full control, run the pipeline locally: fandc520/dsh-comfyui exposes ComfyUI workflows as agent tools and dsh-comfyui-canvas gives them a visual IDE — your GPU, your models, no per-image billing. Keep the directions straight: generation is output; reading images (vision) is input, and a different theme.
In this collection
Editor’s picks lead, then the rest of the theme is auto-discovered by GitHub topics and ranked by stars. Catalog data, refreshed on every sync.
- 1vox-directorby alisa0808
Turn one topic into a finished Vox-style paper-collage explainer/ad video — automated end to end on Atlas Cloud + ffmpeg. An agent skill.
Editor’s pickVoice & Speech2,024309 - 2dsh-image-genby shanliuling
AI image studio for DeepSeek Harness — generate, edit & compare images in chat, with 500+ prompts, gallery, multi-model workflows and ComfyUI.
Editor’s pickCuratedDesign, Media & Vision47233 - 3dsh-imagegenby dickpy
DSH (DeepSeek Harness) Web GUI AI image generation plugin: text-to-image & image-to-image via OpenAI-compatible endpoints (gpt-image-2), with shared cross-device history.
Editor’s pickCuratedDesign, Media & Vision8813 - 4dsh-comfyuiby fandc520
Agent tools comfyui_run/comfyui_object_info/comfyui_workflow submit and inspect ComfyUI workflows, render results in-chat.
Editor’s pickCuratedDesign, Media & Vision8311 - 5ark-cliby volcengine
The fastest way to put Volcengine Ark in your terminal and your AI agent — go from prompt to generated media, multimodal answer, or deployed endpoint in a single command, no API glue code.
CuratedMCP & Protocols13814 - 6dsh-directorxby laplaceyoung
DirectorX as a DeepSeek Harness plugin: AI video/image/audio skills, knowledge corpus, and configurable vision/image/video/audio model tools.
CuratedAgents, Automation & Workflows299 - 7dsh-media-skillsby mjorgin
Free image reading & generation for DeepSeek Harness (rc.7 / rc.8 / v0.1.1-rc.1 / rc.2 / v0.1.2-alpha.3) — paste-image reading with auto vision transcription, DeepSeek-V4-Flash-Vision-Exp / GLM-4V-Flash / SenseNova / Gemini failover, Kolors + U1 Fast generation. No keys in repo.
CuratedDesign, Media & Vision182 - 8dsh-short-video-studioby fengyungithub
Adds image/video generation, canvas workspaces, ComfyUI workflow registration, media assets, and scene-production skills to DSH.
CuratedDesign, Media & Vision112 - 9atlas-cloud-skillsby atlascloudai
Atlas Cloud skills for Claude Code, Codex & Gemini CLI — generate images/videos and call 300+ AI models from your coding agent.
MCP & Protocols333 - 10dsh-game-material-masterby universe-st
Generates game sprites, images, and sequence frames through Seedream/MiniMax with local ffmpeg keying, extraction, and compositing.
CuratedDesign, Media & Vision141 - 11mcp-serverby atlascloudai
MCP server for Atlas Cloud - AI API aggregation platform for image/video generation and LLM
MCP & Protocols113 - 12dsh-agnes-studioby zmm863-commits
AI image and video studio as a floating DSH panel, so creation runs alongside the conversation instead of replacing it: text-to-image, image-to-image and multi-image composition at 1K-4K across eight aspect ratios; text-to-video and image-to-video with first-frame control at 4-12 seconds; short-drama mode that imports a script (.txt/.md/.json), breaks it into storyboard shots for preview and batch generation; and a prompt-expert workspace. Zero runtime dependencies.
CuratedDesign, Media & Vision22
Frequently asked questions
How do I generate images by chatting in DSH?
Install shanliuling/dsh-image-gen and describe what you want in the conversation — images come back inline, follow-up edits and side-by-side comparisons work in the same thread, and a 500+ prompt library ships with Seedream, Gemini and OpenAI routes switchable in settings.
Do image generation plugins need an API key or a GPU?
Depends on the route. Cloud-route plugins (dsh-image-gen, dsh-imagegen, vox-director) only need an API key in settings — no GPU on your machine. Local-route plugins (dsh-comfyui, dsh-comfyui-canvas) drive your own ComfyUI: you bring the GPU and models, and skip per-image billing.
Can these plugins generate videos too?
Yes. alisa0808/vox-director (★2,024) builds a complete Vox-style explainer video from one topic; synmindai/dsh-seedance2 and malenia12/seedance-video-generator hook up Seedance video models; yu-wenchao/dsh-tool-lipsync makes lip-synced talking clips. A dedicated walkthrough is coming to /learn/video-generation.
Image generation vs. image recognition plugins — what's the difference?
Opposite directions. Generation is output: the agent turns text into images and video. Recognition is input: the agent reads the screenshots, photos and documents you give it (vision, OCR, structured extraction) — that is the vision-multimodal theme, covered in /learn/image-recognition. They can coexist, but each plugin points one way.
Explore related categories
Explore the full directory
Hundreds of plugins across every category — or submit your own and get listed.