Плагины
Находите, фильтруйте и устанавливайте плагины DeepSeek-Harness.
Найдено плагинов: 193
dsh-nebula-spectrum
liuhongxumaritime/dsh-nebula-spectrum
Aurora-canvas theme in eight editions — Indigo, Ember, Crimson and Neon (dark), Rainbow, Sunrise, Glacier and Rose (light) — that re-points the DSH design tokens so the canvas, glass sidebar, spectrum-rimmed composer, gradient send button and code blocks follow the palette, with a settings row to switch editions or follow the system appearance.
gis-workbench
yinchuanqi/gis-workbench
MapLibre GIS workbench for DSH: a shared map canvas with layer management (visibility, ordering, removal), raster and vector import (SHP, GeoJSON, GeoPackage, ZIP), semantic styling and building labels, and a chat-callable gis_add_layer tool. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-yingma-canvas
barnetshen/dsh-yingma-canvas
Opens the Xiaoyingma AI video canvas in a DSH sidebar and dock tab; agents read and edit canvas nodes through the bundled yingma CLI and MCP tools. Needs a Xiaoyingma account. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-wallpaper-studio
xiaojiuop/dsh-wallpaper-studio
Paints an uploaded image, a video file, a direct URL, or one of 6 built-in gradients behind the whole DSH interface, with fill/fit/tile scaling, wallpaper opacity, background blur and a darkening veil. Overrides the primary, canvas and panel design tokens, keeping separate light and dark values, and sets independent opacity for panels, inputs and popovers. Media stays in the browser (IndexedDB) and is never uploaded.
dsh-wechat-skin
mingtian-lii/dsh-wechat-skin
WeChat-style theme for the DSH Web GUI: green accents and send action, green user bubbles, a soft gray chat canvas, a green sidebar whale, a badge on the current session, a context-usage status dot, and a prominent compaction card.
dsh-english-lit-reader
yubococo/dsh-english-lit-reader
English paper reader inside DeepSeek Harness: drag in a PDF and select a sentence to get its translation in the card column beside the page, with the model called only when you press Translate. Produces a six-section Chinese analysis (overview, problem and prior methods, technical approach, experimental setup and results, limitations, reproducibility), reading the whole paper in one call when the selected model can fit it and otherwise taking complete excerpt notes first. Resolves the bibliography in three stages: the original entries appear immediately, then Chinese titles, then links found through Crossref and OpenAlex, where a DOI or arXiv id printed in the entry is used directly, "not found" is written instead of a guessed link, and rate-limited lookups are reported separately as temporarily failed with a retry. Explains a selected figure together with its nearby caption when the model accepts images, typesets and explains a formula through KaTeX, explains a selected term against its own sentence and the paper text and keeps it as a glossary entry, records highlights and annotations as two independent kinds that do not overwrite each other, and searches inside the document. Nine interface tones (four light, five dark) plus seven paper colours are applied as a darken-blended layer above the page canvas, so only the white sheet is recoloured and text, figures and tables keep their own colours. Function cards can be reordered, detached and resized; one, two and three page layouts, zoom with Ctrl+wheel, two-way panning, fullscreen, and a per-paper recent list kept in IndexedDB. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
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.
chubby-fish-studio
food211/chubby-fish-studio
A pigment-mixing painting canvas in the right sidebar, where the agent sees the picture, paints on it stroke by stroke, and reads exact pixel colours.
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-workflow-orchestrator
czyt1988/dsh-workflow-orchestrator
Visual workflow orchestrator: drag task nodes on a canvas, wire them into sequential, parallel or gateway-joined graphs, and run them; every task node creates a real DSH task session, with judge and terminate nodes that can end a run early.
influence-net
jqlong17/influence-net
影响网 — A-share influence analysis for DeepSeek Harness (dsh): local engine + sidebar canvas
dsh-ai-fastvideo
veillessrose/dsh-ai-fastvideo
Text-driven video generator for DSH: writes one Canvas page per scene from 341 bundled style prompts, renders frames with a headless browser, then adds TTS narration, music and sound effects with ffmpeg.
dsh-figgenie-paper-diagram
field209/dsh-figgenie-paper-diagram
Draw paper figures as SVG the way top venues draw them, lint them against layout rules, edit them on a sidebar canvas, and export vector PDF or editable draw.io. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-openfilm
howie79794-sys/dsh-openfilm
OpenFilm AI video making, enhanced for DeepSeek Harness — embedded Studio, local MP4 rendering, canvas pointing, render jobs, and a sidechat with screenshots and voice.
dsh-theme-paper-warm
factyy163/dsh-theme-paper-warm
Eye-care warm paper theme for the DSH Web GUI: overrides canvas, surface, text, border and scrollbar tokens to a warm palette (canvas #F2F0E5, ink #100F0F) and adds a toggle row in Settings → General.
dsh-theme-obsidian
inh-ui/dsh-theme-obsidian
Desktop-style client theme for DeepSeek Harness: translucent slate sidebar, #151517 main canvas and #2c2c2e composer dock, four palette presets (obsidian-origin, glacier-midnight, titanium-noir, sonoma-frost), a settings panel for blur, sidebar opacity, ambient glow and content radius, and three GPU performance tiers (ultra, balanced, eco).
dsh-obsidian-skills
stevenjj33/dsh-obsidian-skills
Agent skills for Obsidian, vendored verbatim from the official repository kepano/obsidian-skills (MIT, Steph Ango). Six skills covering Obsidian Flavored Markdown (wikilinks, embeds, callouts, properties), Bases (.base), JSON Canvas (.canvas), the Obsidian CLI, Defuddle and Knap. Documentation only: no CLI binaries are bundled, so obsidian/defuddle/knap are installed separately by the user.
connectors (deepseek-harness)
dotv-ai/connectors
Drive the DOTV infinite canvas from DeepSeek Harness: the agent lists your canvases and their nodes, adds text nodes, and runs image/video/audio generations — over DOTV's public API, with no runtime dependencies and no build step. It reuses the `dotv` CLI login on the machine, or takes an API key in the plugin settings. Generations consume credits on your account, exactly as they do in the app.
dsh-Engram
yhprime/dsh-engram
Experience and skill engine with a sidebar console and exp_record / exp_recall / exp_feedback / exp_consolidate / exp_type tools, stored in a local JSON file. Optional Git backup (off until you configure a remote) pushes the library to a remote you choose; GitHub sign-in uses a device code or a token kept in local settings.
dsh-arknights-splash
18518182595-cell/dsh-arknights-splash
Arknights-style procedural opening film and silent looping chat backdrop for the DeepSeek Harness Web UI: Canvas-drawn visuals, synthesized WebAudio sound, click-anywhere skip, cross-dissolve into the interface, live-adjustable opacity; zero dependencies and no media files. By default the film (about 12 seconds, with sound) plays on every page load; a setting turns it off.
dsh-intertidal
winditer/dsh-intertidal
Adds a full-screen sea overlay that appears on a schedule (3 minutes every 120 minutes by default) and fades out on its own. The phase, sky colours and moon come from the solar position computed on the host from a city table, so the light matches the real time and place. The water is either drawn procedurally on a canvas from a perspective noise field (random patches that grow toward the viewer, with sun glitter, near-field whitecaps and daytime gulls; at night a moon) or taken from images and videos in a folder you choose, or from web links fetched once into a local cache and graded to the current light. Includes quiet hours, deferring while you type, skip-today, a draggable corner-window form and an optional synthesised surf sound. No telemetry; the only network requests are the links you paste.
dsh-plugins (mind-canvas)
feahter/dsh-plugins
Infinite-canvas mind map in the sidebar: grow nodes with Tab, tidy a branch into a tree, ask the assistant to expand a node into child ideas, and export to Markdown, OPML, FreeMind, JSON or PNG.
dsh-model-effort-chips
dreamaniax/dsh-model-effort-chips
Splits the composer model seat into a model chip and a reasoning-effort chip: the effort chip opens a draggable, wheel-adjustable slider, and the panel behind it animates 36 canvas effects across 12 categories.
dsh-memory-cloud
faruzancn/dsh-memory-cloud
Visualizes all DSH session memories as an interactive 3D nebula, with a neuron-signal skin, memory-scaled clusters, zoom-driven detail levels and a live write/read activity stream over SSE.