DeepSeek Harness Plugin Hub

CATEGORY

Models & Usage DSH plugins

Verified manifests and exact versions in this category.

All plugins

876 plugins
v1.0.2
dsh-vision-toggle

Vision Toggle

DSH model vision toggle: adds a “Model Vision” row to the settings page for switching the input vision modality of manually declared models under llm-pi-ai custom routes via the official settings.mutate channel, with changes taking effect immediately.

vision-mediamodels-usage00
Updated 9/3/2026
v1.3.0
@dshp-inx/vision-bridge

Vision Bridge

DeepSeek Harness (DSH) vision bridge: let text-only models delegate image understanding to multimodal vision models via vision_describe tool, with persistent settings UI and primary/fallback auto-retry. 纯文本模型通过 vision_describe 工具把图片转交给多模态视觉模型识别的桥接插件。

vision-mediaintegrations-communicationmodels-usage00
Updated 9/3/2026
v1.0.0
@dshp-inx/token-quota

Token Quota

DSH quota plugin (Token Quota) — supports multi-provider rolling quotas and usage-based balances, real-time sidebar cards, provider management on the settings page, and the local standard Bundle format.

models-usageui-customization00
Updated 9/3/2026
v0.1.4
dsh-llm-providers-ui

Llm Providers Ui

Shared LLM Providers settings shell, card order, and picker sort for DeepSeek Harness plugins

ui-customizationmodels-usage00
Updated 9/3/2026
v0.1.2
dsh-composer-tokens

Composer Tokens

DSH web GUI composer real-time token counter (client plugin, v1)

models-usage00
Updated 9/3/2026
v0.2.1
dsh-gemini-oauth-bridge

Gemini Oauth Bridge

Bridge Google AI subscription (Antigravity/Gemini OAuth, Code Assist API) into DeepSeek Harness as an OpenAI-compatible model endpoint

integrations-communicationsecurity-accessmodels-usage00
Updated 9/3/2026
v0.1.3
dsh-llm-xai-oauth

Llm Xai Oauth

Native SuperGrok / X Premium OAuth provider for DeepSeek Harness. Reuses local grok-bridge tokens; no xAI API key.

models-usagesecurity-access00
Updated 9/3/2026
v0.1.0
dsh-vision-patch

Vision Patch

Per-model and per-route image-input (vision) checkboxes on the Models page's custom-provider cards, writing through the llm-pi-ai settings namespace.

vision-mediamodels-usage00
Updated 9/3/2026
v1.1.0
@dshp-inx/token-stats

Token Stats

DeepSeek Harness (DSH) statistics plugin: aggregates every local session log into cumulative/peak/streak token stats, daily & weekly trends, hour-of-day distribution, a GitHub-style year heatmap, and per-model breakdown, shown in a Settings page section themed with DSH tokens.

models-usage00
Updated 9/3/2026
v0.4.0
dsh-subagent-router

Subagent Router

Model-routed subagent delegation for DeepSeek Harness: a `subagent_model` tool with per-call provider/model/max_tokens overrides and a built-in `model: "auto"` routing policy (anchored to the parent's own model by default; task-tier upgrades, failure escalation, auditable reason), plus a `subagent_m

agents-orchestrationmodels-usage00
Updated 9/3/2026
v0.3.0
dsh-v-token-insight

V Token Insight

V Token Insight · Token Insights — DSH Web session token statistics plugin: third tab (KPIs/charts/per-turn details) + lightweight usage display at the end of replies + cross-session token statistics overview (sidebar entry point, all-session KPIs by workspace/session/time/model + local incremental

models-usageui-customization00
Updated 9/2/2026
v0.1.0
dsh-session-slm-router

Session Slm Router

Shadow-mode SLM router for DeepSeek Harness: per-turn weak/strong prediction via vertical-small-model CLI, writes ~/.dsh/slm-shadow/session-slm-shadow.jsonl; does NOT change the active model (S1 shadow only).

models-usage00
Updated 9/2/2026
v0.2.22
dsh-token-usage-xc

Token Usage Xc

DSH web plugin: Token usage statistics for today and the last 7 days, broken down by model, with a header summary, cache hit rate, and 7-day trend chart.

models-usageui-customization00
Updated 9/2/2026
v0.1.1
@goodandready/dsh-usage-guard

Usage Guard

Keeps a malformed token-usage sample from taking a whole session history down with it: a missing or non-numeric counter becomes zero before the harness adds it up, so the fold never yields NaN and already-broken sessions open again.

models-usagememory-context00
Updated 9/2/2026
v0.1.0
@zhourenke/dsh-tool-call-limit

Tool Call Limit

Per-agent per-step ToolRuntime call limits for DeepSeek Harness

agents-orchestrationmodels-usage00
Updated 9/2/2026
v0.2.0
@xfqz86/dsh-usage-stats

Usage Stats

DSH web plugin: token usage statistics in the sidebar footer — today badge (wide/rail) opening a detail panel with per-model/provider breakdown, session list, daily trend curve and heatmap (host scans session logs, serves /usage-stats/api/snapshot)

models-usageui-customization00
Updated 9/2/2026
v0.1.0
dsh-token-saver

Token Saver

DeepSeek Harness plugin: aggressive token optimization without affecting task quality. Trims tool results, manages context window pressure, optional LLM-summary compaction, stable prompt-cache layout.

models-usage00
Updated 9/2/2026
v0.2.2
dsh-openrouter-free

Openrouter Free

OpenRouter free model panel: automatically fetches a live list of free models (overall capability rated 1–5★ and sorted with the best first), installs them into the openrouter route and switches the current conversation model with one click (writes both the global default and session-level settings;

models-usage00
Updated 9/2/2026
v0.4.0
dsh-provider-pro

Provider Pro

Custom provider enhancements: User-Agent override, reasoning-level switching, and image-input declaration for DSH custom providers · 自定义供应方增强:UA 覆盖、推理等级切换、图片输入声明。

models-usage00
Updated 9/2/2026
v0.1.0
dsh-local-telemetry

Local Telemetry

Local-first Harness runtime telemetry plugin: records request, model, tool, and plugin lifecycle metrics (latency, tokens, cost, errors, and cache), with content collection disabled by default. Local-first Harness telemetry plugin for DeepSeek Harness. npm package dsh-local-telemetry (the dsh-teleme

models-usage00
Updated 9/2/2026
v0.3.3
dsh-subagent-pool

Subagent Pool

DSH named-subagent pool: Independently configurable presets, models, and reasoning levels for sub-agents, with descriptions that tell the main agent when to call each sub-agent

agents-orchestrationmodels-usage00
Updated 9/2/2026
v0.1.0
dsh-llm-verifier-pro

Llm Verifier Pro

LLM-as-a-Verifier for DeepSeek Harness (unified): fine-grained reward tools (verify_compare / verify_select / verify_track) with Probabilistic Pivot Tournament and vLLM/SGLang prefill, PLUS the Best-of-N conversation mode with a Web settings panel. Merged from dsh-llm-as-a-verifier (TaurenMountain)

models-usage00
Updated 9/2/2026
v0.6.0
dsh-llm-error-retry

Llm Error Retry

DSH (DeepSeek Harness) plugin: Automatically retries failed LLM requests after sleeping according to configured rules. Supports three matching patterns: HTTP status codes, machine codes, and provider response field=value pairs. Each rule has independently configurable sleep duration and maximum retr

models-usage00
Updated 9/2/2026
v1.0.0
dsh-vision-assist

Vision Assist

DeepSeek Harness vision assistant plugin: equips models without vision capabilities with a switchable multimodal recognition model. Images in the input field are automatically saved to disk and rewritten as text prompts; the main model can view images by calling the vision_recognize tool. The recogn

vision-mediamodels-usage00
Updated 9/2/2026
v0.1.0
dsh-usage-export

Usage Export

Usage cost, transcript export, and expanded stats for DeepSeek Harness.

models-usageproductivity-workflow00
Updated 9/2/2026
v0.1.0
dsh-ai-wallet

Ai Wallet

DSH plugin: one-click account balance & quota panel for multiple AI providers (DeepSeek, Zhipu GLM, Z.ai, Moonshot, MiniMax, StepFun, SiliconFlow, OpenRouter, xAI).

models-usage00
Updated 9/2/2026
v0.1.0-preview.2
dsh-provider-passport

Provider Passport

Review-first request-dialect preflight for custom DeepSeek Harness providers.

models-usage00
Updated 9/2/2026
v1.0.0
dsh-kimi-provider

Kimi Provider

Kimi (Kimi Code account) integration bundle for DeepSeek Harness: provider route, models, device-code login and auto token refresh (npm: dsh-kimi-provider)

models-usagesecurity-access00
Updated 9/2/2026
v0.1.6
dsh-grok-oauth

Grok Oauth

Grok (xAI) subscription provider for DeepSeek Harness: PKCE login, chat, usage chip, and optional Grok Imagine.

models-usagesecurity-access00
Updated 9/2/2026
v0.2.0
dsh-model-compat-guard

Model Compat Guard

DSH compatibility guard: repairs auto-compaction failures for GPT/reasoning models (raises the summary token cap and lowers reasoning effort on purpose=compaction requests), strips doomed sandbox-escalation requests, and auto-fills the required tool-call 'description' argument other models often omi

memory-contextsecurity-accessmodels-usage00
Updated 9/1/2026