DeepSeek Harness (DSH) 插件:在对话中提供实时 iOS Simulator 与 USB 连接的 iPhone。22 个 agent 工具,支持启动、构建、通过可访问性标识/OCR 文本/列表行驱动 UI,并附带可点击和拖拽的流式侧边栏面板。DeepSeek Harness (DSH) plugin: a live iOS Simulator — and a USB-connected iPhone — inside the conversation. 22 agent tools for booting, building, driving the UI by accessibility identity, OCR text or list rows, plus a streaming sidebar panel you can tap and drag on.
仓库/Skill 库
1420 个 · Agent 智能体
一份配置即可统管所有 AI agents:跨项目跨会话可移植、内置精选写作/路由/skills、更安全(内置破坏性命令防护)One config to rule all your AI agents: portable (every project, every session), effective (curated writing, routing, skills), and safer (destructive-command guard).
AgenticX 是一个统一、生产就绪的多 Agent 平台——Python SDK + CLI (agx) + Studio server + Machi 桌面应用。具备 Meta-Agent 编排、15+ LLM provider、MCP Hub、分层记忆、头像与群聊、Skill 生态、安全沙箱和 IM 网关(飞书/微信)。AgenticX is a unified, production-ready multi-agent platform — Python SDK + CLI (agx) + Studio server + Machi desktop app. Features Meta-Agent orchestration, 15+ LLM providers, MCP Hub, hierarchical memory, avatar & group chat, skill ecosystem, safety sandbox, and IM gateway (Feishu/WeChat).
面向组织级 Agent 的开源控制平面与运行时:共享公司上下文、隔离执行、审批流与 MCP。Open-source control plane and runtime for organisational agents: shared company context, isolated execution, approvals and MCP.
面向研究工作流的通用 SKILL.md 目录:覆盖文献综述、研究设计、项目记忆、稿件写作与跨 Agent 委托,兼容 Claude Code、Codex、Gemini、Cursor、OpenClaw 及通用 AI 客户端Universal SKILL.md catalog for research workflows: literature review, research design, project memory, manuscript writing, and cross-agent delegation for Claude Code, Codex, Gemini, Cursor, OpenClaw, and generic AI clients.
🔥🔥🔥 企业级 AI 中间件,可替代 unifyapps、n8n、lyzr🔥🔥🔥 Enterprise AI middleware, alternative to unifyapps, n8n, lyzr
DeepSeek Harness (dsh) 从 0 到 1 深度手册:安装/插件开发/性能调优/实测案例/同模型多 Agent 实测对比(中文 + 英文 PDF)
Opptrix——AI 驱动的全球多市场投研工作台 | 面向中国 A 股的开源 LLM 投研助手。170+ MCP 工具、因子筛选、回测、自选股与 Electron 桌面端。TypeScript · React · Fastify monorepo。Opptrix — AI驱动的全球多市场投研工作台 | Open-source LLM research assistant for China A-shares. 170+ MCP tools, factor screening, backtest, watchlist & Electron desktop. TypeScript · React · Fastify monorepo.
[ICML 2026] effGen:让小型语言模型具备自主 Agent 能力[ICML 2026] effGen: Enabling Small Language Models as Capable Autonomous Agents
Model Context Protocol (MCP) 服务器的 Node.js 实现,支持从多源(包括 **Web of Science**、arXiv 等)搜索和下载学术论文。A Node.js implementation of the Model Context Protocol (MCP) server for searching and downloading academic papers from multiple sources, including **Web of Science**, arXiv, and more.
纯指令包技能,封装 Ai2 的 Asta MCP 服务器(Semantic Scholar)。支持意图到工具的路由、安全默认配置及多 Agent。Pure instruction-pack skill wrapping Ai2's Asta MCP server (Semantic Scholar). Intent-to-tool routing, safe defaults, multi-agent.
pgbot——面向 AI Agent 与应用的 Postgres 智能层pgbot - Postgres intelligence for AI agents & apps
DOI → PDF 解析器,7 源回退(Unpaywall、S2、arXiv、PMC、bioRxiv、出版商、Sci-Hub)。支持多 Agent,零依赖 Python。DOI → PDF resolver with 7-source fallback (Unpaywall, S2, arXiv, PMC, bioRxiv, publisher, Sci-Hub). Multi-agent, zero-deps Python.
AI 驱动的 CLI 编码 Agent,内置 20+ 工具,支持 MCP 与多模型提供商。AI-powered CLI coding agent with 20+ built-in tools, MCP support, and multi-model providers
DSH 官方尚无终端 TUI 的补位之作:Claude Code 风格全屏交互终端插件——像素鲸鱼顶栏、实时工作状态行、思考流式展开、双击 Esc 回滚、上下文进度条 + TPS 仪表。npm 一键安装。
面向 user-aware Agent 的开源记忆与上下文方案:作用域化记忆、来源追溯、检索质量、纠错、边界、评测,以及 MCP/HTTP 访问能力。Open-source memory and context for user-aware agents: scoped memory, provenance, retrieval quality, correction, boundaries, evals, and MCP/HTTP access.
Semantic Scholar API 的 FastMCP 服务器实现,提供学术论文数据、作者信息及引用网络的全面访问。A FastMCP server implementation for the Semantic Scholar API, providing comprehensive access to academic paper data, author information, and citation networks.
开源办公 Agent 与游戏平台 Beta 2026。Open Source Office Agent & Game Platform Beta 2026
面向 SolCex Exchange 的自主 AI BD Agent:7×24 跨链代币评分与支付 2026。Autonomous AI BD Agent for SolCex Exchange: 24/7 Cross-Chain Token Scoring & Payments 2026
一款实验性编辑器,提供与 AI 协作编写软件的新方式。An experimental editor that offers a novel way to write software with AI.
AKB — Agent Knowledgebase。AI agent 的组织记忆:以 URI 图统一 vault 范围的文档 / 表格 / 文件,通过 MCP 提供服务。AKB — Agent Knowledgebase. Organizational memory for AI agents: vault-scoped docs / tables / files unified by URI graph, served over MCP.
Unica (Ю́ника)——面向 1С:Предприятие 开发的 Codex 与 Claude Code 公开插件。Unica (Ю́ника) — публичный плагин Codex и Claude Code для разработки на 1С:Предприятии.
本地法语语音助手:支持 Claude 或 Ollama(离线)、Hue 智能家居、OBS、日历、浏览器、Twilio 通话与 MCP server。基于 Python。Assistant vocal local en francais : Claude ou Ollama (offline), domotique Hue, OBS, agenda, navigateur, appels Twilio, serveur MCP. Python.
自托管运行时,用于将受控的项目工具暴露给 ChatGPT GPT Actions 与 MCP 客户端。Self-hosted runtime for exposing controlled project tools to ChatGPT GPT Actions and MCP clients.
303 份 AI/LLM 中文讲义,支持在线阅读、PDF 下载和 LaTeX 源码查看 | Stanford CS336/CS224R/CS25 | Berkeley LLM Agents | Agent 工程实践
面向 AI Agent(Claude Code、Codex、OpenClaw、Hermes)的 Skill:清除俄语文本中的 37 类 AI 生成痕迹(官腔、ChatGPT 套话、复制粘贴残留),并检测文本是否由 AI 生成。俄语 AI 文本人性化与检测 Skill,专为编程 Agent 设计Skill для AI-агентов (Claude Code, Codex, OpenClaw, Hermes): убирает 37 признаков AI-генерации из русского текста и проверяет, писала ли его нейросеть - канцелярит, штампы ChatGPT, артефакты копипаста | Russian AI-text humanizer & detector skill for coding agents
由 AI Pulse Georgia 策划的尖端 AI agent 框架、开发者工具与自动化资源精选集A curated collection of cutting-edge AI agent frameworks, developer tools, and automation resources — by AI Pulse Georgia
以 Skill 形式实现的学术文献发现 — 支持 Claude Code · Codex · 任何能加载 SKILL.md 的 Agent。五大数据源 · 四级分层 · 单文件 Shadcn 报告。Academic literature discovery as a Skill — Claude Code · Codex · any agent that loads SKILL.md. Five sources · four tiers · single-file Shadcn report.
一个全栈 AI 驱动应用,自动抓取、摘要、聚类并格式化研究论文为结构化文献综述——为研究者节省大量人工时间。A full-stack AI-powered application that automatically fetches, summarizes, clusters, and formats research papers into a structured literature review — saving researchers hours of manual effort.
DeepSeek Harness 的任务感知推理模式路由器:三个已实测的行为带(spec/mixed/react),附带相变证据,persona + 首轮工具注入,agent 可见的可调参数。包含双吸引子策略论文。Task-aware reasoning-mode router for DeepSeek Harness: three measured behavior bands (spec/mixed/react) with phase-transition evidence, persona + first-turn tool injection, agent-visible tuning. Dual-attractor policy paper included.
围绕代码 Agent 的前沿研究活体合集,从我们构建的代码到我们行动的世界。A living collection of frontier research on code agents, from the code we build to the worlds we act in.