插件

Agentic Control Plane for DeepSeek Harness — check every tool call against your policies before it runs, and keep a durable record of what was allowed and why.

Expose DeepSeek Harness agent capabilities as an MCP server, so an external MCP client (e.g. Hermes) can drive Harness to execute coding tasks. Hermes = brain, Harness = arms.

OpenAI Codex OAuth provider and native settings UI for DeepSeek Harness

Connect DeepSeek Harness agents to authenticated Codex App Server sessions

Declarative Claude Code-style permission rules for DeepSeek Harness: ordered allow/deny/ask rules with tool-name, argument (glob/regex), and workspace-path matching on the tools/pre-execute waterfall, full session-log audit, and HMR rule reload.

Run Pi extensions as agent-scoped DeepSeek Harness plugins

Structured long-term memory system for DeepSeek Harness -- unit-based brief/full pairs with auto-injection

DeepSeek Harness (DSH) 个人微信 ClawBot 插件：OpenClaw 负责微信通道（扫码登录/收发消息），DSH 负责大脑，两者通过 OpenAI 兼容协议桥接。

DeepSeek Harness 外置视觉模型插件：inspect_image 把本地图片或 http(s) 图片 URL 发给任意 OpenAI 兼容端点，视觉模型看图的文字回答直接带回对话；附 Web UI 设置栏。

DeepSeek Harness 的 TencentDB Agent Memory 移植：L0 对话捕获 → L1 结构化记忆提取 → L2 场景/L3 画像，自动召回注入 + 记忆/对话搜索工具；复用现有 ~/.memory-tencentdb/memory-tdai 数据；附 Web UI 设置栏。

A Symphony-compatible multi-provider task orchestrator and dashboard for DeepSeek Harness

Lightweight requirement alignment for DeepSeek Harness — align important decisions before execution without imposing a full spec workflow.

A model-native meta-agent plugin for DeepSeek Harness: the model synthesizes task-specific workflows at runtime from five agentic workflow patterns (prompt chaining, parallel workers, router, supervisor, evaluation loop) and coordinates tools and subagent

Bridge OpenAI Codex's memory, skills, and configuration into DeepSeek Harness

Langfuse observability for DeepSeek Harness: exports session logs as OpenTelemetry traces (GenAI semantic conventions) to Langfuse's OTLP endpoint

Stable, editable PPTX generation for AI agents — semantic IR in, native DrawingML out

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。

Codex 风格桌面宠物：右下角悬浮动画精灵，随 agent 运行状态实时变化（工作、等待、报错、完成）。

复用 Codex CLI 的 ChatGPT 登录态注册 `openai-codex` LLM 路由，并在 DSH Web 设置中提供 GPT Auth 控件。

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。

SHOPLINE AI 工具包：接入 SHOPLINE 官方开发者 MCP 服务器，内置七个 SHOPLINE 平台技能（Admin REST、GraphQL、OAuth、Webhook、Sline），是 Shopify AI Toolkit 的 SHOPLINE 对应版。

Markdown 笔记的双知识图谱 RAG：前置依赖 + 相似关联双图（P/S-Agent 遍历）、段落级匹配、ngram/embedding 混合检索、显式权重反馈闭环，以 MCP 接入。

把 Markdown 渲染为自包含的独立 HTML 页面：提供在 headless 配置下同样可用的 `md_html_render` 工具，以及在网页端浏览、预览、编辑并导出本地 `.md` 文件的抽屉；两个入口共用同一个渲染器，无运行时依赖。

基于代码库知识图谱的 read_graph 工具（CONTAINS / EXPORTS / IMPORTS / IMPORTS_SYMBOL 关系）。