DeepSeek Harness Plugin Hub

Publish and manage complete Harness Profiles. Discover Plugins for your next setup.

Explore

PluginsPresetsDocsNews

Community

Publish a pluginContactReport an issue

Resources

Plugin Hub on GitHubDeepSeek HarnessSystem statusPrivacy notice
© 2026 DeepSeek Harness Plugin HubPowered byPaxTech

Independent and unofficial. Not affiliated with, authorized by, or endorsed by DeepSeek.

Models & Usage DSH plugins — Page 34

CATEGORY

Models & Usage DSH plugins

Verified manifests and exact versions in this category.

All categoriesAgents & Orchestration1035Memory & Context1312Coding & Developer Tools2813UI & Customization3666Integrations & Communication1920Vision & Media617Search & Research866Security & Access1264Models & Usage1681Productivity & Workflow2252

All plugins

1687 plugins
v1.0.5
dsh-slot-health

Slot Health

Live health pane for local LLM endpoints (llama.cpp slots + Ollama models) in the DeepSeek Harness web UI.

ui-customizationmodels-usage★ 0↓ 0
Updated 10/15/2026
Sv3.0.0
dsh-subagent-role-router

Subagent Role Router

Route DSH subagent delegations to configured models: 3-role routing (explore / plan / patch), round-robin or random distribution, cross-provider failover, fork window slicing, and agent-teams exemption.

agents-orchestrationmodels-usage★ 0↓ 0
Updated 10/15/2026
Hv0.2.0-rc.33
hanamesh-usage

Hanamesh Usage

Consent-gated, device-signed HanaMesh usage events for DSH

models-usagesecurity-access★ 0↓ 0
Updated 10/15/2026
Lv0.2.3
dsh-llm-moe4all

Llm Moe4all

MoE4All local inference plugin for DeepSeek Harness, with engine management, model discovery, and downloads.

models-usage★ 0↓ 0
Updated 10/15/2026
Mv0.9.0
dsh-model-organizer

Model Organizer

按厂商与模型系列浏览，支持折叠、全局搜索、名称/倍率排序、置顶、别名与隐藏。 | Group models by provider and family, search globally, sort by name or multiplier, and manage pins, aliases and visibility.

models-usage★ 0↓ 0
Updated 10/15/2026
Gv0.1.0
dsh-gemini-kitten-widget

Gemini Kitten Widget

A Gemini catgirl quota widget in the bottom-right corner of the DSH interface: displays today’s used/remaining requests on the free tier of the Google Gemini API (20 requests per model per day, reset at midnight UTC), with model switching, quota-based color warnings, drag-and-snap positioning, and c

models-usageui-customization★ 0↓ 0
Updated 10/15/2026
Fv0.7.17
dsh-freellmapi

Freellmapi

Self-contained free-tier LLM routing for DeepSeek Harness: one `auto` model that scores each request's complexity and picks the best of 18 catalogued providers -- five enabled by default -- failing over automatically. Unconfigured providers collapse to a single 'add your key' entry instead of dead m

models-usage★ 0↓ 0
Updated 10/15/2026
Pv0.1.1
dsh-plugin-1tempo

Plugin 1tempo

MiL-Bridge: convert a request into MiL with Sarashina, review it, then hand it to the model dsh is running (TUI and Web UI)

models-usageintegrations-communication★ 0↓ 0
Updated 10/15/2026
v0.21.4
dsh-model

Model

Many model sources behind one authenticated OpenAI-compatible endpoint for DeepSeek Harness (dsh) and your tools: WorkBuddy, OpenCode Zen and your CLI subscriptions (Codex, Kimi…), with a management page inside dsh. 多家模型汇成一个 OpenAI 兼容地址，dsh 和其他软件都能直接用。

models-usagesecurity-access★ 0↓ 0
Updated 10/15/2026
Mv1.0.0
dsh-model-icons

Model Icons

Vendor marks in the DeepSeek Harness model picker: the host half mirrors the provider's model catalog and logo set to loopback, and the browser half draws one monochrome mark per model row, the way the provider's own dashboard draws it.

models-usageui-customization★ 0↓ 0
Updated 10/15/2026
Cv1.0.0
dsh-cc-switch

Cc Switch

把 cc-switch 里配好的模型供应商直接接入 DeepSeek Harness：读取 cc-switch 数据库，转换成 llm-pi-ai 路由并写入凭据，还能实测每条端点通不通。Bridges your cc-switch providers into DeepSeek Harness as llm-pi-ai routes.

models-usageintegrations-communication★ 0↓ 0
Updated 10/15/2026
Pv0.2.0
dsh-plugin-session-cost

Plugin Session Cost

Shows what the current session has cost, plus the signed-in DeepSeek account balance, in the composer dock.

models-usage★ 0↓ 0
Updated 10/15/2026
Pv0.1.1
dsh-plugin-token-bucket

Plugin Token Bucket

A little bucket that drains while your DeepSeek API balance is being spent.

models-usage★ 0↓ 0
Updated 10/15/2026
Xv1.4.2
dsh-xl-free-model

Xl Free Model

XL Free Model — Install this plugin in dsh to automatically run a local OpenAI-compatible proxy and use free models such as mimo-v2.6-flash-free through the OpenCode free channel; the models are registered directly in the session model selection list (provider: xl-free-model), with no need to manual

models-usage★ 0↓ 0
Updated 10/15/2026
Mv0.1.0
dsh-model-switchboard

Model Switchboard

Manage all models in DSH from one place: the Settings page’s “Model Toggles” lists every provider and model, including those dynamically registered by plugins and not visible in the built-in Models page. Toggle individual models or entire groups; disabled models disappear from all model selectors. A

models-usage↓ 0
Updated 10/15/2026
Pv0.1.1
dsh-peakclock

Peakclock

Peak/off-peak countdown: time periods and countdowns in model names, with real-time status, display preview, and billing details.

models-usageui-customization★ 0↓ 0
Updated 10/15/2026
Wv1.0.0
dsh-whale-pixel-pet

Whale Pixel Pet

像素鲸鱼娘桌宠 for DeepSeek Harness: An interactive companion that stays in the GUI, changes poses based on session activity, and grows with Token usage

ui-customizationmodels-usage★ 0↓ 0
Updated 10/15/2026
Ov0.1.2
@foreveryu/dsh-opencode-go

Opencode Go

OpenCode Go and Zen live model discovery, session routing, pricing and eligibility diagnostics for DeepSeek Harness

models-usagedeveloper-tools★ 0↓ 0
Updated 10/15/2026
Sv0.1.0
dsh-standardcompute

Standardcompute

Standard Compute provider for DeepSeek Harness: one OpenAI-compatible route with a cost router and direct model pins.

models-usage★ 0↓ 0
Updated 10/15/2026
v1.0.0
dsh-strata-monitor

Strata Monitor

Shows the Strata model server's monitor panel (state, speed, GPU, cache, requests) inside the Harness UI.

models-usageui-customization★ 0↓ 0
Updated 10/15/2026
v0.1.4
dsh-model-kit

Model Kit

Model parameter editing, imports, and native USD/CNY pricing for DeepSeek Harness.

models-usage★ 0↓ 0
Updated 10/15/2026
Mv1.2.0
dsh-medic-pet

Medic Pet

Local-usage desk pet for DeepSeek Harness: an original chibi field-medic mascot (hand-drawn SVG, no official assets) that reports per-turn token usage, grows one step per day of real use, and adds local stats, daily goals, badges, sit/water reminders and outfit themes. No network, no credentials.

models-usageproductivity-workflow★ 0↓ 0
Updated 10/15/2026
Pv1.0.0
dsh-plugin-team-model-routing

Plugin Team Model Routing

Let an Agent Teams Lead choose the provider, model, and reasoning effort of each teammate it spawns.

models-usageagents-orchestration★ 0↓ 0
Updated 10/15/2026
Lv0.1.0
dsh-litellm-extra-headers

Litellm Extra Headers

DeepSeek Harness plugin: inject an `extra_headers` entry into the OpenAI-compatible request BODY for model calls routed to listed providers and sent to listed upstream hosts — the only injection point that survives a LiteLLM relay, which drops client HTTP headers.

models-usage★ 0↓ 0
Updated 10/15/2026
Qv0.1.9
dsh-quota-pill

Quota Pill

Shows the DeepSeek account balance, the current Session's credit spend, and peak/off-peak pricing beside the shipped composer stats.

models-usageui-customization★ 0↓ 0
Updated 10/15/2026
Tv1.0.0
@local/token-breakdown-stats

Token Breakdown Stats

Session token usage with subagent rollup: replaces the composer-dock stats pill so the bottom total covers the main agent plus every subagent it called, and the click-open dialog itemizes usage by bucket, by subagent, and by model.

models-usage★ 0↓ 0
Updated 10/15/2026
Pv0.1.0
dsh-pylon-trae

Pylon Trae

Trae LLM provider for DeepSeek Harness: reads the credential an already signed-in Trae desktop or CLI install wrote, serves it over a fixed loopback shim, and seeds one editable provider entry on the models page. Trae's own data is read-only.

models-usagesecurity-access★ 0↓ 0
Updated 10/15/2026
Pv0.4.0
dsh-pylon-workbuddy

Pylon Workbuddy

DeepSeek Harness plugin: supervise a local workbuddy2api gateway process and write its models into an editable llm-pi-ai provider

models-usage★ 0↓ 0
Updated 10/15/2026
Sv0.0.1
dsh-stream-index-guard

Stream Index Guard

Patches the LLM assistant-stream accumulator at runtime to clamp invalid chunk indices, downgrading hard stream failures to dropped or merged chunks; ships a stats tool.

developer-toolsmodels-usage★ 0↓ 0
Updated 10/15/2026
Tv0.0.1
dsh-ttft-hedge

Ttft Hedge

Hedged requests for streaming /v1/messages LLM calls: fires a second attempt when response headers miss a deadline and races them, retries immediately on 5xx/504, passes through 4xx/429, with statistics.

models-usage★ 0↓ 0
Updated 10/15/2026
← Previous1…3233343536…57Next →
DeepSeek Harness Plugin Hub
ProfilesPluginsCategoriesNewsDocsSign inManage Profiles
ProfilesPluginsCategoriesNewsDocsSign in