Categories › Models & AI
Models & AI
868 pluginsModel access, vision, voice, agents and personas — extending what DSH can do.
Sorted by downloads. 100 per page.
dsh-openrouter-livefails
OpenRouter provider that reads the live /models list, so new models appear without a harne
↓ 1.6k
dsh-qq-accesstested
Personal QQ channel for dsh: managed NapCat with QR login, private/group AI replies with b
↓ 1.6k
dsh-vision-webtested
Vision for text-only DeepSeek via Doubao Web by default (zero-cost, no API key — drives yo
↓ 1.6k
dsh-dseyesfails
Native image attachments for text-only DeepSeek in the Web GUI: pasted or dropped images a
↓ 1.6k
dsh-help-me-to-asktested
On the first real work message of each chat window, asks for the missing goal, deliverable
↓ 1.6k
dsh-spark-scopetested
Shows a Spark Scope card at the bottom of the dsh sidebar: model server decode and prefill
↓ 1.6k
@martindye/dsh-voicetested
Two-way voice for the DSH web app: dictate with the unified composer, hear replies with sp
↓ 1.6k
dsh-codex-authfails
Reuses the Codex CLI ChatGPT login as an `openai-codex` LLM route and adds GPT Auth contro
↓ 1.6k
dsh-interim-narrationtested
Makes the agent narrate its process: every turn opens with how it reads the message and wh
↓ 1.6k
@flowingspring/dsh-vocotested
Continuous voice conversations for DSH with hands-free listening, push-to-talk, speech rec
↓ 1.6k
dsh-request-paramstested
Add your own parameters to the model requests dsh sends: OpenRouter provider routing and f
↓ 1.6k
dsh-comfyui-imagetested
Image generation through a local ComfyUI server: one generate_image tool covering text2img
↓ 1.6k
dsh-experttested
Expert roster for DeepSeek Harness: 636 experts across 22 divisions with full Chinese name
↓ 1.6k
dsh-model-kittested
Fills missing model parameters from models.dev, imports models from configured providers,
↓ 1.6k
dsh-voice-announcerfails
Voice announcements when a conversation ends (session title, round, outcome) plus live sen
↓ 1.5k
dsh-persona-forgetested
Rewrites your composer draft in the voice of a chosen character card before it is sent, us
↓ 1.5k
dsh-omni-workstationtested
Omni-modal workstation for DSH: analyze_image over an ordered multi-card VLM failover chai
↓ 1.5k
dsh-plugin-rollout-scouttested
Detects which conversation model an account is being served: launches concurrent throwaway
↓ 1.5k
dsh-connect-trae-modelstested
Aggregates all local Trae accounts into one auto-failover model provider, with manual and
↓ 1.5k
dsh-design-qafails
Design-fidelity QA for text-only models: a `deepseek_vision` tool borrows an eye from any
↓ 1.5k
dsh-ai-gateway-providertested
Connect several OpenAI-compatible gateways at once: each one is its own group in the model
↓ 1.5k
dsh-model-foldtested
Groups the DSH model picker by provider: the menu lists sources with their model counts, a
↓ 1.5k
dsh-plugin-subagent-rolestested
File-defined subagent roles for DeepSeek Harness: roles live in project and global .dsh/ro
↓ 1.5k
dsh-moa-nativetested
Native generation-time Mixture-of-Agents: every turn, reference models run the same sessio
↓ 1.5k
@max-null/dsh-capturetested
Dual-engine screen capture for DSH: inside the SSiD desktop shell, a global hotkey or tray
↓ 1.5k
dsh-unlimited-retrytested
Retries a model request that no other recovery layer claimed, unbounded by default with an
↓ 1.5k
dsh-antigravity-authfails
Adds an unofficial Antigravity OAuth-backed LLM route, web search, image generation and ed
↓ 1.5k
dsh-cloudflare-mediatested
Cloudflare Workers AI media tools for dsh - text-to-image, text-to-speech and speech-to-te
↓ 1.5k
dsh-tool-visual-primitivestested
Analyzes conversation images through mode-specific prompts (caption, UI, document, groundi
↓ 1.5k
@duke-dsh-plugins/dsh-model-prompt-injectortested
Per-model system prompt injector for DeepSeek Harness: append custom prompt text to the EN
↓ 1.4k
dsh-advisor-grouptested
Consult multiple expert advisor models in a retro CRT chat-group card: one ask_advisors ca
↓ 1.4k
dsh-um-proxy-ondemandtested
Manages the lifecycle of the local UM GenAI proxy (127.0.0.1:8788) that DSH's um-genai pro
↓ 1.4k
dsh-subscription-logintested
One settings page that lists every subscription sign-in the harness can offer and walks yo
↓ 1.4k
dsh-orbital-agentsfails
Persistent Endeavour and Challenger peer sessions for DeepSeek Harness that plan, execute
↓ 1.4k
dsh-image-tilertested
Shows oversized images to the model at readable resolution instead of one downscaled copy:
↓ 1.4k
dsh-opencode-connecttested
Brings the free models of a locally installed OpenCode CLI into DSH: starts or reuses a lo
↓ 1.4k
@benrong/dsh-prompt-polishtested
Turns a one-line DSH draft into a complete task brief (role / goal / constraints / output
↓ 1.4k
dsh-codex-modefails
Codex-shaped coding tools and agent presets for GPT models on DeepSeek Harness: exec_comma
↓ 1.4k
@edge-sky/dsh-oauth-adaptertested
Provides an OAuth account page for DSH Web, currently supporting OpenAI Codex, GitHub Copi
↓ 1.4k
dsh-modellixfails
Modellix integration for DSH: schema-driven image, video and audio generation in a split D
↓ 1.4k
dsh-model-http-proxytested
Only the domains you list go through a proxy: add a model host that needs the proxy to be
↓ 1.4k
dsh-subagent-hubtested
Named subagent presets bound to model-route allowlists, where an out-of-list child model i
↓ 1.4k
dsh-connect-doubaotested
Reuses the sign-in already held by the local Doubao desktop app as a DeepSeek Harness mode
↓ 1.3k
@lutisure/cyrene-dsh-pluginunregistered
A Cyrene-themed agent for DSH, with a customizable look (palette, wallpaper, particle effe
↓ 1.3k
dsh-image2-tooltested
Registers an `image2_generate` tool that calls your own OpenAI-compatible image relay to g
↓ 1.3k
dsh-delegate-routertested
Automatic Flash/Pro routing for DeepSeek Harness subagent calls: light tasks run on a chea
↓ 1.3k
dsh-newapi-providertested
New API gateway provider: several gateways with one active at a time, per-gateway model li
↓ 1.3k
@biliye/dsh-voice-calltested
Voice call assistant for the DSH desktop app and the Web GUI (the same plugin, only the pr
↓ 1.3k
@mangoc/dsh-sub2api-synctested
Syncs the groups, keys and models of one sub2api instance into dsh-llm-pi-ai provider rout
↓ 1.3k
dsh-language-guardtested
Locks the model's reasoning, replies and tool-call descriptions to a chosen target languag
↓ 1.3k
dsh-xai-oauthtested
Sign in to an xAI Grok subscription with device-code OAuth, refresh the access token, and
↓ 1.3k
dsh-subagent-modelstested
Set the provider, model and reasoning effort delegated subagents run with, from one chip b
↓ 1.3k
dsh-saytested
Speaks your agent's reports in a voice you choose, compressing long reports before speakin
↓ 1.3k
dsh-volume-knobtested
Speaker button left of the model selector in the composer tool row — one click scrolls to
↓ 1.3k
dsh-plugin-notifyfails
Notification outbox: agent proactively notifies via toast / Chinese TTS voice / sound effe
↓ 1.3k
dsh-cyber-im-avatartested
IM stand-in seat for DeepSeek Harness: a sidebar panel polls Feishu/Lark, WeCom and DingTa
↓ 1.3k
dsh-darumatested
Keeps long DSH sessions alive when a model channel starts failing: daruma takes over after
↓ 1.2k
dsh-image-litetested
Image generation for DSH in two modes — an image_generate_lite tool the model calls from c
↓ 1.2k
dsh-peakratetested
Peak / off-peak rate badges for the model selector (every row) and the composer tool row,
↓ 1.2k
@superfish058/dsh-llm-proxytested
Per-model proxy routing for LLM requests: selected models go through a proxy (Clash etc.)
↓ 1.2k
dsh-chatvoicetested
Free voice closed loop for the Web UI: browser SpeechRecognition mic input with live inter
↓ 1.2k
dsh-opencode-go-secondtested
Adds an independent OpenCode Go connection with credential settings, manual model discover
↓ 1.2k
dsh-workbuddy2api-paneltested
Reads a running workbuddy2api-panel over its HTTP API and syncs its models into the DSH mo
↓ 1.2k
dsh-opendesign-gotested
OpenDesign Go provider: 12 subscription models over Chat Completions, the real 5-hour and
↓ 1.2k
@lr611/visionforgetested
Vision understanding + image generation/editing for DeepSeek Harness (read images with Qwe
↓ 1.2k
↓ 1.2k
dsh-vision-recognizertested
Vision provider route that transcribes attached images to text through a configurable mode
↓ 1.2k
dsh-vision-mixfails
Combine text, vision, and image-generation APIs into one Mix model with automatic routing:
↓ 1.2k
dsh-quicksighttested
Two-tier image reading for text-only models: fast local OCR (RapidOCR, offline) first, vis
↓ 1.2k
dsh-rotakittested
Pools every WorkBuddy account signed in on this machine into one auto-failing-over provide
↓ 1.2k
@skylerfee/dsh-llm-opencode-go-livetested
An OpenCode Go model source for DSH: it reads the opencode-go catalog from Models.dev and
↓ 1.2k
dsh-youreyestested
Vision toolkit for text-only DeepSeek: model-invokable `vision` tool, wrapper adapters for
↓ 1.2k
dsh-llm-qwen-localtested
LLM adapter plugin for locally deployed Qwen models behind a vLLM OpenAI-compatible endpoi
↓ 1.2k
dsh-as-mcp-servertested
Exposes DSH as an MCP server so another agent can watch its thinking, tool calls and file
↓ 1.2k
dsh-zcode-startplantested
Connects the free ZCode plan to the model picker: a multi-account pool with rotation, per-
↓ 1.2k
dsh-antigravityfails
Google Antigravity / Cloud Code Assist model provider with quota management and model sele
↓ 1.2k
dsh-auto-visiontested
Auto-discovery vision bridge for text-only DeepSeek Harness agents: automatically finds an
↓ 1.1k
dsh-cliproxyapi-protested
Connects CLIProxyAPI (CPA) to DeepSeek Harness (DSH) with multiple accounts, model filteri
↓ 1.1k
dsh-grok-byapitested
Grok model provider for DeepSeek Harness with model discovery, image and video generation,
↓ 1.1k
dsh-proxy-routestested
Per-provider (account) proxy routing for DSH. SOCKS5/HTTP proxy pool with account-level ro
↓ 1.1k
dsh-plugin-team-subagent-controltested
Adds per-member model and reasoning-effort routing to DeepSeek Desktop Agent Teams, plus l
↓ 1.1k
dsh-antigravity-oauthfails
Unofficial Gemini Cloud Code Assist OAuth integration with isolated account storage; carri
↓ 1.1k
dsh-koboldcpp-handsfails
Hands repetitive text and vision labor (OCR, image analysis, comparison) to a local Kobold
↓ 1.1k
dsh-model-pricingtested
Model pricing for the DSH settings page: ~7,250 models / 213 providers with the harness ca
↓ 1.1k
dsh-workbuddy-experttested
Brings WorkBuddy desktop-app experts into DSH: a settings-page market scans the local Work
↓ 1.1k
dsh-glm-modetested
GLM Mode agent preset for zhipuai/glm-5.3-flash: PTC/Code Mode tool presentation, a reason
↓ 1.1k
dsh-model-metadatatested
DSH plugin: matches a model you declared on a custom gateway to the capability metadata it
↓ 1.1k
dsh-hold-to-talktested
One-handed, keyboard-free input for the composer: press and hold the mouse on the input bo
↓ 1.1k
dsh-openrouter-provider-advisorfails
Ranks OpenRouter providers for the current DSH model by quantization, speed, cache-heavy p
↓ 1.1k
@godd6366/dsh-sub2apifails
Connect a sub2api gateway to DeepSeek Harness: OpenAI-compatible multi-provider routes (Op
↓ 1.1k
dsh-plugin-mynahtested
Give the agent a talking face: seven tools that drive a self-hosted Mynah realtime digital
↓ 1.1k
dsh-lanterntested
Expose every model already configured in your local DeepSeek Harness to the LAN over OpenA
↓ 1.1k
dsh-qqbot-muutested
Standalone QQ-bot extension pack for DeepSeek Harness — fully independent, no other plugin
↓ 1.0k
dsh-voice-webspeechfails
Browser Web Speech API voice input: zero server, zero keys, zero model downloads (Edge=Azu
↓ 1.0k
dsh-outsourcing-experttested
Agent presets whose leader does no work itself: every non-management tool call is denied a
↓ 1.0k
@shuind/dsh-codex-harnesstested
Compact Codex coding mode for DSH: light prompt guardrails against GPT over-explaining, wi
↓ 1.0k
dsh-agy-safefails
Compliantly use your own Antigravity quota in dsh: bridges your local logged-in official A
↓ 1.0k
dsh-advisor-pluginfails
Pairs the running executor with a stronger reviewer model through zero-parameter advisor()
↓ 1.0k
@ai-galaxy/dsh-soundtested
Per-event sound notifications: turn completion, approval, question, plan-review, goal-bloc
↓ 1.0k
dsh-models-radartested
Displays CodexRadar model capability benchmarks in a Settings page and shows the selected
↓ 1.0k