跳过主要内容

插件

浏览、筛选并安装 DeepSeek-Harness 插件。

共 222 个插件

J

dsh-token-panel

juhe291/dsh-token-panel

A corner HUD for DeepSeek Harness that shows your session's token pressure, per-model cost, and daily/monthly usage at a glance, with an editable budget and balance that tracks spending for you. 右下角常驻的 Token 仪表盘:实时看会话压力、按模型估算花费。预算和余额点一下就能改,每天每月用了多少都有记录。

4前天UI 增强MIT
J

dsh-stream-rules

jiesou/dsh-stream-rules

Inject rules when needed, without wasting context. DSH port of opencode-stream-rules.

4前天工具与能力MIT
H

dsh-codex-provider

hu9956/dsh-codex-provider

OpenAI Codex provider for DeepSeek Harness with device-code OAuth, Codex CLI import, token refresh, and a web settings panel.

4前天UI 增强MIT
D

dsh-openai-oauth

dgpisces/dsh-openai-oauth

DeepSeek Harness LLM provider for Codex models using managed ChatGPT OAuth

4前天模型与账号接入MIT
D

deepseek-harness-openai-oauth

dgpisces/deepseek-harness-openai-oauth

DeepSeek Harness LLM provider for Codex models using managed ChatGPT OAuth

4前天模型与账号接入MIT
0

qiushi-dsh-evidence-audit

030611/qiushi-dsh-evidence-audit

把工具结果与会话事件的 receipt 写入本地哈希链 JSONL,不保存提示词、工具参数、结果正文或原始会话 ID。

4前天开发与运行时MIT
B

dsh-testgen

bujue600-arch/dsh-testgen

自动化单元测试生成:/testgen 命令与 generate_tests 工具,生成、运行并修复测试直至通过(LLM 与离线模板双生成器;支持 vitest/jest/mocha/node:test)。

4前天开发与运行时MIT
W

dsh-feishu-bridge

wz-heng/dsh-feishu-bridge

Fail-closed 的飞书(Lark)channel 桥:和机器人聊天即驱动 dsh agent turn。仅经官方 Python SDK 集成(精确锁版);白名单默认全拒、webhook 验签/时间窗/防重放、按 chat 粘性会话;中英双语文档。

4前天通知与集成MIT
Z

dsh-continual-evolve

zk-andy/dsh-continual-evolve

持续自进化:从会话轨迹沉淀版本化、可审计、可回滚的 harness 状态(提示词/记忆/技能/子代理规格),带审查门禁与技能热加载。

4昨天工具与能力MIT
S

godot-bridge

smalldy/godot-bridge

DSH↔Godot 引擎运行时控制桥:通过游戏内置 TCP 交互服务器启动并操控运行中的 Godot 4.x——8 个工具(场景/UI 检查、GDScript eval、输入模拟、截图、headless 静态操作、脚本编译检查),替代 godot-mcp。

4前天MCP 与连接器MIT
M

dsh-usage-chart

max-samson/dsh-usage-chart

输入框下方的用量/成本/余额仪表盘:实时指标指示器 + 零依赖 SVG 用量图表,按轮次统计用量、估算成本并实时查询 DeepSeek 账户余额。

4昨天UI 增强MIT
L

dsh-web-attention-badge

luaphes/dsh-web-attention-badge

会话需要你时三处同时亮起:角标、标签页标题计数、按状态换色的鲸鱼 favicon。

4前天UI 增强MIT
G

dsh-opencode-go-quota

glfzr/dsh-opencode-go-quota

模型选择器左侧的 OpenCode Go 额度圆环:点击循环切换 5 小时/每周/每月用量窗口,按紧急程度着色(绿/蓝/橙/红),悬停显示百分比与重置倒计时。

4昨天UI 增强MIT
L

dsh-mcp-lens

labmimors/dsh-mcp-lens

渐进披露 MCP 网关:用 `mcp_search` 检索大型远程工具目录,再由 `mcp_call` 按精确 schema 调用,并采用惰性连接与有界缓存。

3昨天开发与运行时MIT
T

dsh-any-background

tkingxiao/dsh-any-background

自定义外观插件:换背景图(可调大小和位置)、用色轮自定义主题色,主界面和设置界面的透明度都能单独调。

3昨天主题与外观MIT
G

dsh-drop-file-to-path

glfzr/dsh-drop-file-to-path

Codex 式拖拽:把任意文件拖入 DSH Web 界面,文件存入 ~/.dsh-dropbox,路径以整块蓝色 chip 插入输入框。

3昨天UI 增强MIT
C

dsh-learn-everything

cendaifeng/dsh-learn-everything

Feynman learning-mode plugin: teach → teach-back → judge → re-explain loop rendered as rich HTML lesson cards (mermaid diagrams + shiki code highlighting).

33天前工具与能力
A

dsh-auto-memory

aik358/dsh-auto-memory

DSH 自动记忆插件:三层记忆(用户级/项目笔记/每日日志)自动注入与检索、每日反思、可视化面板与设置页,支持继承其他 AI 工具的历史记忆。An auto-memory plugin for the DeepSeek Harness Web GUI: three-layer memory (user-level / project notes / daily logs) with automatic injection and retrieval, daily reflections, a visual panel and settings page, and inheritance of memories from other AI tools.

3昨天UI 增强BSD-3-Clause
J

dsh-reasoning-settings

junelearn/dsh-reasoning-settings

让 DeepSeek Harness 的第三方 API 支持低、中、高等推理强度,并可为每次子 Agent 调用选择模型|Add Low, Medium, High, and other reasoning levels to third-party APIs, with model selection for each subagent call

3前天UI 增强MIT
D

dsh-lineage

dongsheng123132/dsh-lineage

Content-addressed artifact, fact, action and report lineage for DeepSeek Harness

3前天工具与能力MIT
H

dsh-report-html

hccccc01333/dsh-report-html

Generate self-contained interactive HTML reports from Markdown, tables, charts, China province maps, flowcharts, math, and drill-down tables — a DeepSeek Harness (dsh) plugin

3前天工具与能力
Y

dsh-blue-whale-maid

yuxino/dsh-blue-whale-maid

运行在 deepseek harness 上的女仆酱

3前天娱乐
O

dsh-daily-progress

omdsh-dev/dsh-daily-progress

33天前工具与能力BSD-3-Clause
S

better-model-provider

sanshanya/better-model-provider

Per-model capability declaration for DeepSeek Harness: reasoning-effort levels (wire spellings) + request modalities (vision) for OpenAI-compatible providers. Settings section, zero runtime harness deps, no YAML.

3昨天模型与账号接入MIT