跳过主要内容

插件

浏览、筛选并安装 DeepSeek-Harness 插件。

共 12 个插件

T

dsh-model-router

tianji-qingtian/dsh-model-router

模型路由与成本优化器:简单问题 flash 直答、故障自动降级、会话 token/缓存/成本实时面板 | Model router & cost optimizer for DeepSeek Harness: flash quick-answers for simple questions, failure fallback, live token/cache/cost panel

8前天UI 增强MIT
V

dsh-llm-fallback

visol-456/dsh-llm-fallback

DeepSeek Harness 回退链插件:主模型失败自动切换备用 provider,带 Web UI 配置面板 | Provider fallback chains for DeepSeek Harness

4前天UI 增强MIT
O

dsh-llm-fallbacks

omdsh-dev/dsh-llm-fallbacks

Automatic provider/model fallback chains for DeepSeek Harness agents when LLM requests keep failing (retry exhausted, auth, quota, rate limit)

3昨天模型与账号接入MIT
B

dsh-llm-fallbacks

btspoony/dsh-llm-fallbacks

基于角色的模型重试与备用策略。

3昨天模型与账号接入MIT
C

dsh-bilibili

czx2244/dsh-bilibili

B站视频分析工具:提取元数据、字幕文稿(必剪/本地 ASR 兜底)、评论与弹幕,抓取清晰关键帧并可选本地视觉描述。

3昨天工具与能力MIT
T

dsh-deeptutor

tecfancy/dsh-deeptutor

Learning assistant extension for DeepSeek Harness (dsh): DeepTutor tutoring for your agent — deep explanations, self-test questions, learning paths, personal knowledge-base search (RAG), and note archiving (HTTP/WS first, CLI fallback; auto-adapts local/r

2昨天开发与运行时
A

dsh-browser

anweat/dsh-browser

自包含浏览器运行时:Playwright(chromium)+ OpenCLI 作为插件本地依赖(全局复用回退),提供 `browser` 服务与 9 个交互式浏览器工具。

2前天娱乐MIT
L

dsh-model-failover

letter2025/dsh-model-failover

两级模型熔断与回退:模型或平台连续失败后自动熔断,并把下一个请求路由到配置好的备用模型。

2前天工作流与自动化
T

dsh-web-search-exa

tonydua/dsh-web-search-exa

ctx.web 接缝的零配置 Exa 网页搜索提供方:无 API key 时走匿名 MCP 兜底,配 key 时走 REST 搜索。

2前天MCP 与连接器MIT
G

dsh-shift-router

green-dalii/dsh-shift-router

Two-tier model router for DeepSeek Harness — LLM-Judge routing, multi-model fallback chains, exponential-backoff failover, and task-level orchestration (DSH adaptation of pi-shift-router)

1昨天工具与能力MIT
J

dsh-polyglot

jesse-njx/dsh-polyglot

DSH 的模型切换器:指向任意 OpenAI 兼容端点,内置精选免费/低价 DeepSeek 服务商预设,免费额度限流时自动回退。

1前天开发与运行时MIT
L

dsh-whale-animation

leemancheung/dsh-whale-animation

DSH Web 状态文字旁的持久化黑色鲸鱼深潜动画,提供减少动态效果回退与无缝闭环。

1前天UI 增强MIT