CATEGORY
Verified manifests and exact versions in this category.
@ponydev/dsh-model-usage
DSH Web 插件：实时显示模型供应商额度与用量（DeepSeek / GLM Coding Plan / OpenRouter / Kimi Coding Plan + 自定义适配器）
dsh-chatgpt-sub
Sign in with your ChatGPT account to use ChatGPT models in DSH.
dsh-agy-safe
dsh plugin: use a logged-in Antigravity CLI (agy) headless session as the chat and subagent model provider
@wingsky-1/dsh-opencode-usage
OpenCode Go plan usage overlay: persistent three-window usage percentages in the top-right corner, expanding into three independent mini charts for rolling/weekly/monthly (each window has an independently adaptive axis, a 100% quota reference line, reset marker line, and background sampling to prese
@y2zyyr/dsh-token-usage-sidebar
Persistent Token Usage insights for DeepSeek Harness, backed by a scalable durable SQLite ledger.
dsh-model-tidy
DeepSeek Harness plugin: keep the model selector tidy. Choose which models appear and what they are called, without touching any provider configuration or running session.
dsh-mimo-connect
将 MiMo 桌面 App 包含的模型自动接入 DeepSeek Harness，在 DSH 对话窗口里零配置使用。复用客户端已登录的额度，无需单独申请 API Key，也无需常驻程序。Bring the models included in the MiMo desktop App into DeepSeek Harness automatically, with zero configuration.
dsh-peak-cost-mode
DeepSeek peak-pricing cost guard for DSH: during peak hours (Beijing 09:00–12:00 / 14:00–18:00, ×2) the agent automatically switches to ultra-compressed caveman-style output to save tokens, and a header badge + toast keep you posted on every peak/valley t
dsh-keys-manager
Multi-key management for DSH providers: sticky per-session round-robin, retry failover, cooldown recovery, and searchable extra providers.
dsh-context-truth
DeepSeek Harness plugin: tells the model its real, host-measured context occupancy every turn — and keeps compaction summaries from carrying false "context is nearly exhausted" claims.
dsh-proxy-routes
DSH (DeepSeek Harness) plugin: per-provider (account) proxy routing with a named proxy pool -?route each LLM account through its own SOCKS5/HTTP(S) proxy via DSH's bundled undici, configurable live from the DSH settings page.
@kidli1412/dsh-token-heatmap
DSH web plugin: GitHub-style daily token-usage heatmap on the new-session screen with switchable year/month views and six color schemes. The card configures itself through a floating ⚙ panel (palette + default view) — no DSH settings card — and shows toda
dsh-modellix
Modellix Design, LLM, and Web providers for DeepSeek Harness
dsh-plugin-llm-balance
DSH (DeepSeek Harness) plugin: a draggable API balance/quota floating card in the top-right corner of the page—automatically discovers providers such as DeepSeek / Moonshot / Kimi For Coding, changes color by balance tier, and refreshes in real time
@sidequest-007/dsh-atlas
AT Last, All Sources. One @ to jump anywhere — line ranges like @src/a.ts:12-40, the token cost of each reference before you send, and Esc to walk the session back.
dsh-api-hub
API Hub — manage and call local API providers (keys, models, balances) and skills from inside DeepSeek Harness
@guojinpeng/dsh-cost-meter
DSH session and per-turn cost estimation plugin, supporting models.dev price synchronization and custom model unit prices. Configuration is located at ~/.dsh/guojinpeng/dsh-cost-meter/config.json.
@adaptiqai/deepseek-harness
AdaptiqAI for DeepSeek Harness: models with 5M-token native memory and Knowledge. Sign up, get a model and work without leaving the harness
dsh-subagent-default-model
Configure default models for DeepSeek Harness sub-agents: sequential/random allocation across multiple models, cross-provider failover on connection failure, display the current provider and model in trajectory/conversation views, and inject the current routing into the sub-agent context with a sing
@freespace8/dsh-deepseek-balance
Official DeepSeek API balance pill for the DeepSeek Harness Web GUI session header, with click/timed refresh and auto-hide for non-official models. 官方 DeepSeek API 余额胶囊。
dsh-verification-receipt
Privacy-minimal heuristic per-turn execution summaries for DeepSeek Harness
@andrecgoncalves/dsh-balance
Provider-aware account balance chip for the DeepSeek Harness web sidebar: DeepSeek, OpenRouter, Moonshot/Kimi, Zhipu/GLM, and MiniMax.
dsh-delegate-router
Delegate light subagent tasks to cheaper models: automatic Flash/Pro routing for DeepSeek Harness subagent calls, with /delegate overrides, budget caps, and a routing-ledger panel.
dsh-cohub
DeepSeek Harness CoHub: Chinese agent orchestration — users can choose between 2 work modes (co-orchestrator pure orchestration / cohub-cordis creative mode) + 13 specialized agent skills + multi-model consensus; additionally, the two supporting presets cohub-subagent-readonly / cohub-subagent-write
dsh-key-switch
DeepSeek Harness plugin: keep several DeepSeek API keys, switch the active one in one click, and see each key's live balance in the sidebar. 多个 API Key 池 + 一键切换 + 实时余额。
dsh-service-monitor
DeepSeek Harness 第三方服务余额监控插件：选择服务、录入 API Key、按 1/5/10/30 分钟定时查询剩余额度，提供主面板看板与侧边栏状态图标。只收录有余额查询接口的服务（Tavily、博查、Firecrawl、SerpAPI、TinyFish）。Third-party service balance monitor for DeepSeek Harness: pick services, enter API keys, refresh quota on a 1/5/10/30-m
dsh-custom-headers
Per-model custom request headers for DeepSeek Harness: define named header profiles in Settings → Plugins → Plugin configuration, pick one per model in the provider's model catalog, and every call to that model carries those headers.
@lynn123411/dsh-a6api
A6API multi-model aggregation platform integration plugin: native LLM provider registration, real-time account balance and price fluctuation monitoring, Token whitelist synchronization, merchant route probing, and comprehensive metric cards
dsh-credits
DeepSeek Harness quota plugin: displays built-in plans, balances, or custom HTTP quotas by provider, and provides model pricing, current-session estimates, and cumulative usage
@ztlovelsw/dsh-model-profile
DSH Web GUI model capability configuration: Directly set reasoning efforts for each configured model in the model catalog editor under “Settings → Models,” and sync reasoning efforts and context capacity in batches or according to models.dev presets, writing them to the llm-pi-ai settings namespace.
