Discover The Best DSH plugins

DeepSeek Harness plugins — Page 290

21674results

Search community plugins for DeepSeek Harness. Compare capabilities, installation instructions and compatibility to find extensions for your AI workflow.

dsh-agent-arcade

A deterministic Agent-played Snake game for DeepSeek Harness.

Agent collaboration
dsh-agent-board

跨 Agent 工作台:左侧栏 Agent 面板 + /agent-view 实时看板 + 继续工作按钮 + 设置页(常驻插件)

Agent collaboration · Developer tools
dsh-agent-bridge

Bridge DeepSeek Harness to whatever coding-agent CLI is installed (OpenCode, Claude Code, Codex, Gemini, ...) through data-driven recipes: auto-discovery, capability-driven UI, per-run reports.

Agent collaboration · Interface extensions · Developer tools · Integrations
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:mike-sl-ig/dsh-agent-bridge
dsh-agent-browser

浏览器工具插件:client 捕获 console/全局错误上报宿主,agent 用 browser_console 工具读取——自己看 F12 Console。

Agent collaboration
dsh-agent-compact-self

自主压缩插件(爱丽丝为核心):session_compact 工具原语——何时压缩由爱丽丝自主决策(框架零强制),调用后经 compaction seam(AgentCompactEngine)注入总结指令,爱丽丝输出 <compacted-summary> checkpoint 完成事务。

Agent collaboration
dsh-agent-context

感知当前会话的上下文占用与已花费 token:/context 命令 + ctx.contextMeter 服务(只读聚合官方 token-meter 与 session-projection)

Agent collaboration · Productivity
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:jonah791/dsh-agent-context
dsh-agent-context-pruner

上下文剪枝插件(爱丽丝判断为核心):候选检测(含缓存代价估算)+ 按需剪枝执行(replay-safe,头+标记+尾)+ 统计。是否无用由爱丽丝判断;剪枝不破坏缓存命中(尾部优先)。

Agent collaboration
dsh-agent-control

DeepSeek Harness 插件:把「删除一个会话」与「从模型上下文中删除一轮对话」这两件破坏性操作收进一个插件,用同一套确认与失败语义管理。

Agent collaboration · Productivity
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:ventisyn/dsh-agent-control
dsh-agent-driver-writehere

Hierarchical long-form writing for DeepSeek Harness — WriteHERE as an agent driver, not extra ReAct tools.

Agent collaboration
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:Player-YN/dsh-agent-driver-writehere
dsh-agent-evolve

跨代自评估进化插件:配置仓库(版本化资源+锚点链)、热重载(.agent-presets 运行时应用/回滚)、子智能体白纸继承配置跑 modeltest 评测、账本(分数/轨迹/失败模式)。受益主体为宿主 agent 本体。

Agent collaboration
dsh-agent-forge

Forge a solidifiable agent preset+plugin from any code workspace, progressively

Agent collaboration · Developer tools · Productivity
dsh-agent-grid

定员八席 —— 把 DeepSeek Harness 的侧栏换成 2×4 的常驻席位看板:八个 agent 各自的实时动作、状态与待办一眼看全。Not affiliated with DeepSeek.

Agent collaboration · Interface extensions
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:yefengliu1/dsh-agent-grid
dsh-agent-hub

智能体协作台 —— 多智能体房间/群聊/派单控制台(DSH 内嵌,可上插件市场)

Agent collaboration
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:jax629321/dsh-agent-hub
dsh-agent-instructions

在 dsh Web 设置页里编辑原生 AGENTS.md 的面板:指令链上每一层都是可点击的编辑入口,三种生效范围(工作区 + 全局 / 仅工作区 / 仅全局)通过给文件名加 .disabled 切换(内容不删),并可视化官方指令预算与分层占用。

Agent collaboration
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:Bay-Zeddie/dsh-agent-instructions
dsh-agent-lang

工具调用描述 / 模型思考 / 回复输出三个语言通道:跟随界面语言、强制指定或关闭,一条运行时指示实现,不改动预设与工具 schema。· Language control for tool-call descriptions, model thinking, and replies: follow the GUI language, force a fixed tag, or off — via one runtime-context directive, no preset or tool schema touched.

Agent collaboration · Interface extensions
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:KannaKuron/dsh-agent-lang
dsh-agent-lead

DSH 带队模式插件:主 Agent 只规划派活监工,改动交给 subagent,每次派活弹窗选子模型

Agent collaboration
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:JunguangJiang/dsh-agent-lead
dsh-agent-life

数字生命循环原语:可打断的 sleep(到期自我唤醒)+ 醒来状态注入 + 事故文件读取 + 决策日志。零规则——睡多久/醒来做什么由 agent 自主决定。

Agent collaboration
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:jonah791/dsh-agent-life
dsh-agent-llm-retry

LLM 运维一体化插件:模型请求自动多次重试(策略升级 maxRetries 20)+ Token 预算跟踪(token_budget_* 工具,合并自 dsh-agent-token-budget)

Agent collaboration
dsh-agent-log

Agent 工作记录 / 任务复盘 / 自动生成项目报告:从 DSH 会话日志只读还原这次任务做了什么,并产出可读的《本次 Agent 工作报告》

Agent collaboration · Productivity
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:Neo65536-engineer/dsh-agent-log
dsh-agent-plugin-manager

插件管理器:插件档案库(清单/用途/工具/配置认知)+ 生命周期管理(创建/挂载/启停/卸载/配置),host 工具面 + 官方设置页「插件管理」tab。

Agent collaboration
dsh-agent-project-sync

Discover Codex and Claude project directories and register them as shared DeepSeek Harness workspaces.

Agent collaboration · Developer tools · Productivity
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:Harzva/dsh-agent-project-sync
dsh-agent-quality-diagnosis

DSH Web plugin for Agent execution quality diagnosis reports.

Skills · Agent collaboration · Developer tools
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:cat552/dsh-agent-quality-diagnosis
dsh-agent-replay

Replay and share redacted DeepSeek Harness sessions as standalone HTML

Agent collaboration · Developer tools · Productivity
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:forrestsweet/dsh-agent-replay
dsh-agent-skill-forge

被动技能熔炉(Trace2Skill 思想落地):后台采集会话轨迹索引(零 LLM 成本)+ 信号送达;蒸馏/合并/剪枝决策归爱丽丝;技能输出为 SKILL.md(DSH 技能目录原生可加载)

Skills · Agent collaboration · Financial data · Productivity

Find the right plugin, then get started

  1. Search by task

    Use a feature keyword or category, then sort by recent updates or GitHub stars.

  2. Explore capabilities

    Read the description, AI analysis and original README to understand what a plugin does.

  3. Check and install

    Review compatibility, dependency ranges and instructions, then copy the available install command.

Questions about DSH Hub

What is DSH Hub?

DSH Hub is a community plugin directory for DeepSeek Harness, bringing together descriptions, source links and installation instructions to help you discover and compare extensions.

Does inclusion mean a plugin is tested and compatible?

No. Detail pages show recorded compatibility and dependency ranges. Unknown does not mean compatible; check the project documentation before installing.

How do I install a plugin?

Open its detail page and check the installation section. Copy the command into your Harness environment when available, or follow the project README.