{"data":{"slug":"jnmetacode-agency-orchestrator","name":"agency-orchestrator","tagline":"🚀 One sentence → multi-AI-role collaboration → complete plan in minutes. Built on the agency-agents role library (216+ experts), zero-code YAML, web Studio + desktop app, 10 LLM providers (7 free). 基","github_url":"https://github.com/jnMetaCode/agency-orchestrator","owner":"jnMetaCode","repo":"agency-orchestrator","owner_avatar_url":"https://avatars.githubusercontent.com/u/12096460?v=4","primary_language":"TypeScript","stars":1790,"forks":238,"topics":["agency-agents","agent-orchestration","ai-agents","autogen-alternative","claude","codex","copilot","crewai-alternative","cursor","deepseek","gemini-cli","langgraph-alternative","mcp-server","multi-agent","no-api-key","ollama","openai","orchestrator","workflow","yaml-workflow"],"archived":false,"github_pushed_at":"2026-07-15T10:39:28+00:00","maintenance_label":"Very active","url":"https://www.graphcanon.com/tools/jnmetacode-agency-orchestrator","markdown_url":"https://www.graphcanon.com/tools/jnmetacode-agency-orchestrator.md","api_url":"https://www.graphcanon.com/api/graphcanon/tools/jnmetacode-agency-orchestrator","graph_url":"https://www.graphcanon.com/api/graphcanon/graph?tool=jnmetacode-agency-orchestrator","description":"🚀 One sentence → multi-AI-role collaboration → complete plan in minutes. Built on the agency-agents role library (216+ experts), zero-code YAML, web Studio + desktop app, 10 LLM providers (7 free). 基于 agency-agents 专家库，一句话调度多个 AI 专家自动协作，几分钟交付完整方案。","homepage_url":"https://ao.aiolaola.com","license":"Apache-2.0","open_issues":14,"watchers":8,"ai_summary":null,"readme_excerpt":"# Agency Orchestrator\n\n**中文** | [English](./README.en.md)\n\n> **一句话，让多个 AI 角色自动协作，几分钟出完整方案。**\n>\n> **也是你的「一人公司」：你当老板，AI 当团队——自动组队、重大决策请你签字、按验收标准交付。**\n\n\n\n\n\n\n**一句话出结果 · 216 个专业 AI 角色 · YAML 零代码 · 11 种大模型 · 支持 key（推荐 DeepSeek），也有 7 种免 key 方式**\n\n> 📖 [完整上手教程](https://mp.weixin.qq.com/s/XcGbkMb6TM6NLQiL7ICwbw) — 从安装到实战，10 分钟上手\n>\n> 📖 **免费配套学习** → [从零学会 AI 编程](https://aiolaola.com/?utm_source=github&utm_campaign=orchestrator)：180 节免费实操课 + 《AI 编程实战三卷书》在线阅读 + 实战社区 · 永久免费\n\n> 觉得有用？请点个 **Star** — 帮助更多人发现这个项目。\n\n

\n \"网页
\n 网页 Studio：打一句话，AI 自动从 200+ 专家里组队并运行\n

\n\n---\n\n## 网页 Studio（图形界面）\n\n不想敲命令行？本地跑一条 `ao web`，浏览器里勾选专家、运行工作流、查看产物、实时介入——全程图形界面，全中英双语。\n\n> 🆕 **「一人公司」系列模板**：做产品 / 做内容 / 做投研 + 全员大会，关键步骤带验收标准（`acceptance`），投研含老板签字闸门——交付的是可验收的工作成果，不承诺奇迹。\n> 🆕 **AI 自动组队**：不知道选哪些专家？角色页一句话、不选角色，AI 自动从全部专家里挑人组队并运行。\n> 🆕 **可视化画布**：工作流可在画布上拖拽节点 / 连线（自动防环）/ 改任务·角色 / 保存，运行时节点按状态实时点亮。\n> 🆕 **创意库**：内置图像生成提示词库（Nano Banana / Gemini，可搜索 / 分类 / 一键复制）。\n> 🆕 **零配置首跑**：本机已登录 Claude Code / Gemini CLI 等？AO 自动探测并直接用，连 API key 都不用配。\n\n

\n \"Studio
\n 工作流：内置模板一键运行，也能对比多个模板\n

\n\n> 启动：`ao web`（本地，密钥只存你自己机器、绝不外传）。也有 [桌面客户端下载](https://github.com/jnMetaCode/agency-orchestrator/releases/latest)（Electron · macOS / Windows / Linux）。\n> 英文界面同样完整 → 见 [English README](./README.en.md)。\n\n---\n\n## 一句话出结果\n\n也可以纯命令行——一条命令，一句话出结果：\n\n```bash\nao compose \"我是一个程序员，想用AI做自媒体副业，目标月入2万，帮我做完整规划\" --run\n```\n\n

\n \"ao\n

\n\n5 个 AI 角色自动分工协作：\n\n```\n 工作流: 程序员AI自媒体副业规划\n 步骤数: 5 | 模型: claude-code\n 参与者: 🔭 趋势研究员 | 📱 平台分析师 | 💰 财务规划师 | ✍️ 内容策略师 | 📋 执行规划师\n──────────────────────────────────────────────────\n\n ✅ 🔭 趋势研究员 31.3s → 6个赛道竞争度/变现天花板/AI提效倍数对比\n ✅ 📱 平台分析师 32.0s → 6大平台三维评分，推荐\"小红书+公众号\"组合\n ✅ 💰 财务规划师 31.8s → 月入2万拆解：课程¥11,880 + 社群¥2,488 + 咨询¥4,000\n ✅ ✍️ 内容策略师 44.6s → 20个选题 + 4套标题模板 + 内容SOP\n ✅ 📋 执行规划师 42.2s → 90天行动计划，精确到每天做什么\n\n==================================================\n 完成: 5/5 步 | 182.1s | 6,493 tokens\n==================================================\n```\n\n**不用写代码，不用写配置，不用选角色。** 一句话 → AI 自动拆解任务 → 从 216 个角色中匹配 → 按 DAG 并行执行 → 输出完整方案。\n\n### 你能用它做什么\n\n```bash\nao compose \"帮我分析做一个AI记账工具的可行性\" --run # 创业可行性分析\nao compose \"对比 Cursor、Windsurf 和 Copilot，给出选择建议\" --run # 技术选型报告\nao compose \"写一篇关于 AI Agent 趋势的深度文章\" --run # 深度长文写作\nao compose \"用 10 万块启动一个 AI 教育项目\" --run # 商业计划书\nao compose \"PR 代码审查，覆盖安全和性能\" --run # 代码审查报告\nao compose \"设计一个 SaaS 产品的定价策略\" --run # 定价分析\n```\n\n每个场景自动匹配不同的 AI 角色组合。\n\n---\n\n## 为什么需要 Agency Orchestrator\n\n跟一个 AI 聊天，它给你一个视角。但做任何决策，你需要产品的视角、技术的视角、财务的视角、营销的视角……\n\n**Agency Orchestrator = 让多个 AI 专家各干各的，最后汇总。相当于一个人 vs 一个团队。**\n\n| | ChatGPT / Claude | CrewAI / LangGraph | **Agency Orchestrator** |\n|---|--------|-----------|---------------------|\n| 角色数 | 1 个通用 | 自己写 | **216 个专业角色** |\n| 使用方式 | 对话 | 写 Python | **一句话 / YAML** |\n| API key | — | 必须 | **支持 key，也有 7 种免 key 方式** |\n| 依赖 | — | pip + 几十个包 | **npm + 2 个依赖** |\n| 并行 | — | 手动建图 | **DAG 自动检测** |\n| 中文角色 | — | 无 | **216 个** |\n| 价格 | 订阅制 | 开源 + API 费 | **DeepSeek 甜区极低成本，亦可免 key 起步** |\n\n## 3 步开始\n\n### 第 1 步：安装\n\n```bash\nnpm install -g agency-orchestrator\n```\n\n> **装 CLI 还是桌面端？**\n> - **桌面客户端**（[下载](https://github.com/jnMetaCode/agency-orchestrator/releases/latest)）**自带引擎与 Node，双击即用，无需** `npm i -g agency-orchestrator`。只有想在终端用 `ao` 命令、或接进脚本 / CI 时才需要装 CLI。\n> - 用 `--provider claude-code`（或 `gemini-cli` / `codex-cli` 等）时，需要**本机已安装并登录对应 CLI**；AO 会自动探测已装的，零配置直接用。用 API key 类（deepseek/openai…）则配好 key 即可，无需装任何 CLI。\n> - **自定义目录**：产物 / 数据目录用 `AO_DATA_DIR`（桌面端默认指向 userData），角色库用 `AO_AGENTS_DIR`，统一工作区用 `AO_HOME`。\n\n### 第 2 步：一句话跑起来\n\n```bash\n# 用你已有的 Claude 会员（无需 API key）\nao compose \"帮我分析做一个AI记账工具的可行性\" --run --provider claude-code\n\n# 或用 DeepSeek（充 10 块跑很久）\nexport DEEPSEEK_API_KEY=\"你的key\"\nao co","github_created_at":"2026-03-21T18:32:02+00:00","created_at":"2026-07-15T11:18:55.173961+00:00","updated_at":"2026-07-15T11:18:58.849607+00:00","categories":[{"slug":"ai-agents","name":"AI Agents","url":"https://www.graphcanon.com/categories/ai-agents","markdown_url":"https://www.graphcanon.com/categories/ai-agents.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/ai-agents"},{"slug":"inference-serving","name":"Inference & Serving","url":"https://www.graphcanon.com/categories/inference-serving","markdown_url":"https://www.graphcanon.com/categories/inference-serving.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/inference-serving"},{"slug":"llm-frameworks","name":"LLM Frameworks","url":"https://www.graphcanon.com/categories/llm-frameworks","markdown_url":"https://www.graphcanon.com/categories/llm-frameworks.md","api_url":"https://www.graphcanon.com/api/graphcanon/categories/llm-frameworks"}],"tags":[{"slug":"agency-agents","name":"agency-agents"},{"slug":"agent-orchestration","name":"agent-orchestration"},{"slug":"ai-agents","name":"ai-agents"},{"slug":"autogen-alternative","name":"autogen-alternative"},{"slug":"claude","name":"claude"},{"slug":"codex","name":"codex"},{"slug":"copilot","name":"copilot"},{"slug":"crewai-alternative","name":"crewai-alternative"}],"trust":{"provenance":{"is_fork":false,"github_id":1188191061,"owner_type":"User","methodology":"github_public_v1","parent_repo":null,"near_duplicate_slugs":[]},"computed_at":"2026-07-15T11:18:56.296Z","maintenance":{"label":"Very active","score":96,"methodology":"github_public_v1","releases_90d":11,"days_since_push":0,"last_release_at":"2026-07-08T07:18:26Z"},"security_summary":{"status":"findings","scanner":"osv@v1","low_count":28,"high_count":0,"last_scan_at":"2026-07-15T11:18:56.729Z","medium_count":0,"scan_profile":"deps","critical_count":0}},"capability_facts":{"mcp":{"source":"package.json:@modelcontextprotocol/*","observed_at":"2026-07-15T11:18:56.094Z","server_manifest":true},"scan":{"source":"repo_scan","observed_at":"2026-07-15T11:18:56.094Z"},"has_cli":{"value":true,"source":"package.json:bin|scripts","observed_at":"2026-07-15T11:18:56.094Z"},"languages":{"value":["typescript","javascript"],"source":"github.language+package.json","observed_at":"2026-07-15T11:18:56.094Z"},"license_spdx":{"value":"Apache-2.0","source":"github.license","observed_at":"2026-07-15T11:18:56.094Z"}}}}