Plugins
Navegue, filtre e instale plugins do DeepSeek-Harness.
96 plugins encontrados
dsh-cosplay-studio
wang-shuanglin/dsh-cosplay-studio
Local ComfyUI cosplay photo sets for Windows with an NVIDIA GPU (8 GB+ VRAM): deploys the ComfyUI Windows portable build and Z-Image-Turbo (8-10 GB of models) with a self-test, then turns a character name, a reference image, or a hot-character pick into a live-action cosplay photo set with base-image and shot-script approval gates.
dsh-super-image
byton-patrica/dsh-super-image
Generates images in parallel with two configurable OpenAI Images-compatible or ComfyUI backends, compares the candidates, and optionally fuses them. Visual comparison needs a vision-capable model in DSH; without one the candidates are listed unscored.
dsh-comfyui-studio
xingyeai23/dsh-comfyui-studio
ComfyUI studio in a DSH sidebar panel: connects to a local or LAN ComfyUI server, picks saved image workflows, cleans and rewrites prompts, edits LoRAs and nodes, generates images, and previews results with copyable seeds.
media-suite
do-nothing/media-suite
Família de plugins de mídia: instala de uma vez os recursos de mídia Qwen e RunningHub e oferece configurações globais entre provedores.
media-suite (media-runninghub)
do-nothing/media-suite
Executa workflows ComfyUI na nuvem RunningHub como pool em segundo plano: envia lotes, distribui dentro do limite de concorrência, consulta cada job, baixa todos os resultados para o disco e desperta a sessão quando o pool esvazia. Um ledger compartilhado registra lotes, custos, níveis de instância, horários de envio/fim e caminhos de cada arquivo; painel ao vivo na barra lateral direita.
media-suite (media-qwen)
do-nothing/media-suite
Ferramentas de mídia Qwen via Alibaba Cloud DashScope: compreensão multimodal de vídeo e áudio, texto para fala com vozes predefinidas e geração de áudio de cenas.
dsh-shengtu
xuhairui-dzz/dsh-shengtu
Gera imagens com vários mecanismos na conversa: ferramenta de agente e painel acoplado ao compositor com interruptor, presets Volcengine Ark, SiliconFlow, compatível com OpenAI ou personalizado e acompanhamento local de gastos. O usuário fornece a chave API do mecanismo escolhido.
dsh-deck-studio (deck-studio)
yuna78/dsh-deck-studio
Área de trabalho de image-deck para DSH conduzida por chat: transforma a intenção em roteiro de storyboard, gera prompts e imagens de IA por página com revisões individuais, oferece nove estilos de design e exporta PDF/PPTX (uma imagem de página inteira por slide).
dsh-image2-plugin
544244857/dsh-image2-plugin
Registra a ferramenta `image2_generate`, que chama seu próprio relay de imagem compatível com OpenAI para gerar ou editar imagens. Texto para imagem via POST /images/generations; edição com referência PNG/JPEG/WebP local via POST /images/edits; aceita b64_json ou URLs de imagens. Salva no diretório de trabalho da sessão e exibe na conversa pela dependência `@wisdoverse/dsh-inline-media-viewer`.
dsh-grok-byapi
ningyuanlee/dsh-grok-byapi
Provedor de modelos Grok por um gateway compatível com OpenAI, com descoberta de modelos, geração de imagens e vídeos, ferramentas de texto para fala e tarefas de mídia locais persistentes.
dsh-image-bridge
kkj-luna/dsh-image-bridge
Oferece geração texto-imagem e imagem-imagem a modelos que não desenham. Compatível com qualquer API de imagem compatível com OpenAI, vários provedores e modelo alternativo automático.
dsh-plugin-agnes-image
zhp282515-wq/dsh-plugin-agnes-image
Adiciona a ferramenta agnes_image para os modelos Agnes Image Flash (2.5, 2.1, 2.0), para texto→imagem, imagem→imagem e composição de várias imagens. Settings gerencia modelo ativo, tamanho e proporção padrão, pasta de saída, API key e histórico.
dsh-tool-comfyui
yuehancn/dsh-tool-comfyui
Gera imagens no chat chamando um conjunto de workers ComfyUI locais ou na LAN, com modelos de workflow, verificações de saúde e controle da fila.
dsh-media-gen
emberwingaviation/dsh-media-gen
Plugin integrado de geração de imagens e vídeos para DSH: catálogo multi provedores (gateway images/chat compatível com OpenAI, DashScope 通义万相, Seedance da Volcengine, Google Veo e tarefas assíncronas /v1/videos no estilo Sora com input_reference para imagem em vídeo). Ferramentas image_generate / generate_video / animate_image, exibição na conversa, galerias e configurações unificadas. Mescla forks de dsh-image-generation e dsh-video-gen.
dsh-imagegen-skill
dingchenhui0618-arch/dsh-imagegen-skill
Registra a Skill imagegen no registro de Skills, sem copiar arquivos manualmente para `$DSH_HOME/skills`. O agente usa o CLI incluído `image_gen.py` com qualquer endpoint GPT Image compatível com OpenAI para gerar imagens por texto, editar imagens de referência e recortar com fundo transparente.
dsh-imgdraw
ninjasln-labs/dsh-imgdraw
Texto para imagem no DeepSeek Harness: ferramenta de modelo `draw_image`, botão 生图 na barra de entrada com janela de prompt (geração assíncrona, grade com 4 resultados, baixar/manter/excluir), rota de imagem `/imgdraw` e histórico persistido. Backend: DashScope wan2.7-image.
dsh-tool-imagegen
pappet/dsh-tool-imagegen
Geração de texto-para-imagem e imagem-para-imagem pela Image API unificada do OpenRouter, com aliases de modelo configuráveis e parâmetros limitados conforme a lista atual de recursos do modelo, imagens de referência, cartão de configurações e resultados exibidos no chat.
oh-my-dsh
llmpolska/oh-my-dsh
Roteamento de modelos em camadas para o DeepSeek Harness: camadas think/build, delegação de visão (o modelo de visão apenas descreve imagens; o modelo de trabalho age), geração de imagens e uma proteção de alto impacto.
beav-deepseek-harness
jamailar/beav-deepseek-harness
Conecta o DSH ao Beav local para operações de IA no Xiaohongshu (RED/RedNote) e em redes sociais: fluxos de pesquisa de temas, conhecimento, copywriting, imagem, áudio e vídeo curto com aprovação e rastreio de artefatos.
dsh-drawcraft
vhow123/dsh-drawcraft
Drawcraft — an image-generation studio in the sidebar panel list, split into three pages that share one result. Prompt page: pick a target dialect, a style and an aspect, write one rough sentence, and a chat model returns an 11-field shot breakdown that the plugin stitches into the syntax that dialect actually accepts — six dialects (natural English, natural Chinese, GPT-Image, Midjourney, Stable Diffusion, nano-banana), 18 style presets, negative prompts, filler-word stripping and local history; an image can also be reverse-engineered into the same fields, and a one-line edit such as light turns to dusk rewrites only the named fields. Image page: text-to-image and image-to-image through a locally installed image-generation plugin or through an OpenAI-compatible endpoint you add in the panel, whose key stays on disk and is never echoed back to the browser; with a subject image attached it posts to /images/edits and carries up to four extra references. Canvas page: an infinite canvas where image, prompt and result nodes are dragged, resized, wired by hand or by generation, with undo and redo, marquee selection, copy and paste, a minimap, pointer-centred wheel zoom and JSON export and import; rendering from a node walks the whole upstream chain so no reference is dropped, one node can be pinned as the subject, and the reference library opens inside the canvas. A reference library of 4,345 Chinese-labelled entries — 1,439 poses with matching depth maps, 1,407 outfits, 1,499 portraits — is searched and faceted by pose system, angle, category, season, hair length and hairstyle, paged 24 at a time, and the entries you pick are injected into the prompt as binding constraints. Renders can pick the image channel and model, one of eight aspect sizes or follow-the-aspect, and quality at 1K/2K/4K; a preview opens into a lightbox with pointer-centred wheel zoom from 20% to 800%, drag-to-pan, double-click reset and one-click save. The prompt tools and the library are also exposed as the optimize_image_prompt, patch_image_prompt and search_reference_gallery agent tools. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-image-gen
adithyanraj03/dsh-image-gen
Multi-provider image generation and editing for DeepSeek Harness (OpenAI-compatible relays with custom /v1/images/generations endpoints, Gemini, Seedream, DashScope, Grok, GLM, local ComfyUI, subscription accounts), an Excalidraw MIT infinite-canvas studio with click-to-place and selection-to-edit references, and a pro viewer with zoom/pan/pinch, region notes and a markup mode that bakes marks into the pixels and sends normalized coordinates to the model.
dsh-api-free-image
apfsds-dog/dsh-api-free-image
A DSH plugin that provides API-free image generation capabilities. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-imagegen-panel
gpta76/dsh-imagegen-panel
Image-generation sidebar panel for DeepSeek Harness: works with any OpenAI-compatible image endpoint, supports reference-image edits, and ships presets for character sheets, turnarounds and expressions.
dsh-prompt-forge
vhow123/dsh-prompt-forge
Image-generation prompt workshop in the sidebar panel list, working in four directions that share one result. Idea to prompt: pick a target dialect, a style and an aspect, write one rough sentence, and a chat model returns an 11-field shot breakdown that the plugin stitches into the syntax that dialect actually accepts — six dialects (natural English, natural Chinese, GPT-Image, Midjourney, Stable Diffusion, nano-banana), 18 style presets, negative prompts, filler-word stripping and local history. Image to prompt: drop a reference image and a vision model reverse-engineers the same fields. Prompt to prompt: type one line such as light turns to dusk and only the named fields are rewritten, everything else is copied verbatim. Prompt to image: the result card, or the panel itself before any optimization has run, renders through a locally installed image-generation plugin or through an OpenAI-compatible endpoint you add in the panel, whose key stays on disk and is never echoed back to the browser; with a main image attached it posts to /images/edits and carries up to four extra references, with only text it posts to /images/generations. A reference library of 4,345 Chinese-labelled entries — 1,439 poses with matching depth maps, 1,407 outfits, 1,499 portraits — is searched and faceted by pose system, angle, category, season, hair length and hairstyle, paged 24 at a time, and the entries you pick are injected into the prompt as binding constraints. Renders can pick the image channel and model, one of eight aspect sizes or follow-the-aspect, and quality at 1K/2K/4K; a preview opens into a lightbox with pointer-centred wheel zoom from 20% to 800%, drag-to-pan, double-click reset and one-click save. The first two directions and the library are also exposed as the optimize_image_prompt, patch_image_prompt and search_reference_gallery agent tools. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.