[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-1247":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":32,"readmeContent":33,"aiSummary":34,"trendingCount":16,"starSnapshotCount":16,"syncStatus":35,"lastSyncTime":36,"discoverSource":37},1247,"agno","agno-agi\u002Fagno","agno-agi","Build, run, and manage agent platforms.","https:\u002F\u002Fdocs.agno.com",null,"Python",40644,5508,235,432,0,21,123,577,99,45,"Apache License 2.0",false,"main",true,[27,28,29,30,31],"agents","ai","ai-agents","developer-tools","python","2026-06-12 02:00:25","\u003Cdiv align=\"center\" id=\"top\">\n  \u003Ca href=\"https:\u002F\u002Fagno.com\">\n    \u003Cpicture>\n      \u003Csource media=\"(prefers-color-scheme: dark)\" srcset=\"https:\u002F\u002Fagno-public.s3.us-east-1.amazonaws.com\u002Fassets\u002Flogo-dark.svg\">\n      \u003Csource media=\"(prefers-color-scheme: light)\" srcset=\"https:\u002F\u002Fagno-public.s3.us-east-1.amazonaws.com\u002Fassets\u002Flogo-light.svg\">\n      \u003Cimg src=\"https:\u002F\u002Fagno-public.s3.us-east-1.amazonaws.com\u002Fassets\u002Flogo-light.svg\" alt=\"Agno\">\n    \u003C\u002Fpicture>\n  \u003C\u002Fa>\n\u003C\u002Fdiv>\n\n\u003Cp align=\"center\">\n  Agno turns agents into production software.\u003Cbr\u002F>\n  Build agents in any framework. Run as a service. Ship to real users.\n\u003C\u002Fp>\n\n\u003Cdiv align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fdocs.agno.com\">Docs\u003C\u002Fa>\n   • \n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fagno-agi\u002Fagno\u002Ftree\u002Fmain\u002Fcookbook\">Cookbook\u003C\u002Fa>\n   • \n  \u003Ca href=\"https:\u002F\u002Fdocs.agno.com\u002Ffirst-agent\">Quickstart\u003C\u002Fa>\n\u003C\u002Fdiv>\n\n## What is Agno\n\nAgno is the runtime for agentic software. Use it to run agents as a production service.\n\nBuild agents using any framework. Run them as production services with sessions, tracing, scheduling, and RBAC. Manage them from a single control plane.\n\n| Layer | What it does |\n|-------|--------------|\n| **SDK** | Build agents, teams, and workflows with memory, knowledge, guardrails, and 100+ integrations. |\n| **Runtime** | Serve agents in production via a stateless, session-scoped FastAPI backend. |\n| **Control Plane** | Test, monitor, and manage your system from the [AgentOS UI](https:\u002F\u002Fos.agno.com). |\n\n## Quick Start\n\nWrap a coding agent and serve it as a production API. Same shape across every framework.\n\n### With the Agno SDK\n\nSave as `workbench.py`:\n\n```python\nfrom agno.agent import Agent\nfrom agno.db.sqlite import SqliteDb\nfrom agno.os import AgentOS\nfrom agno.tools.workspace import Workspace\n\nworkbench = Agent(\n    name=\"Workbench\",\n    model=\"openai:gpt-5.4\",\n    tools=[Workspace(\".\",\n        allowed=[\"read\", \"list\", \"search\"],\n        confirm=[\"write\", \"edit\", \"delete\", \"shell\"],\n    )],\n    enable_agentic_memory=True,\n    add_history_to_context=True,\n    num_history_runs=3,\n)\n\n# Serve via AgentOS → streaming, auth, session isolation, API endpoints\nagent_os = AgentOS(agents=[workbench], tracing=True, db=SqliteDb(db_file=\"agno.db\"))\napp = agent_os.get_app()\n```\n\n`Workspace(\".\")` scopes the agent to the current directory. `read`, `list`, and `search` run freely; `write`, `edit`, `move`, `delete`, and `shell` require human approval.\n\n### With the Claude Agent SDK\n\n```python\nfrom agno.agents.claude import ClaudeAgent\nfrom agno.db.sqlite import SqliteDb\nfrom agno.os import AgentOS\n\nagent = ClaudeAgent(\n    name=\"Claude Agent\",\n    model=\"claude-opus-4-7\",\n    allowed_tools=[\"Read\", \"Bash\"],\n    permission_mode=\"acceptEdits\",\n)\n\nagent_os = AgentOS(agents=[agent], db=SqliteDb(db_file=\"agno.db\"), tracing=True)\napp = agent_os.get_app()\n```\n\nThe same wrapping pattern works for [LangGraph](#) and [DSPy](#).\n\n\u003Cdetails>\n\u003Csummary>\u003Cstrong>LangGraph\u003C\u002Fstrong>\u003C\u002Fsummary>\n\n```python\nfrom agno.agents.langgraph import LangGraphAgent\nfrom agno.db.sqlite import SqliteDb\nfrom agno.os import AgentOS\nfrom langchain_openai import ChatOpenAI\nfrom langgraph.graph import MessagesState, StateGraph\n\ndef chatbot(state: MessagesState):\n    return {\"messages\": [ChatOpenAI(model=\"gpt-5.4\").invoke(state[\"messages\"])]}\n\ngraph = StateGraph(MessagesState)\ngraph.add_node(\"chatbot\", chatbot)\ngraph.set_entry_point(\"chatbot\")\n\nagent = LangGraphAgent(name=\"LangGraph Chatbot\", graph=graph.compile())\nagent_os = AgentOS(agents=[agent], db=SqliteDb(db_file=\"agno.db\"), tracing=True)\napp = agent_os.get_app()\n```\n\n\u003C\u002Fdetails>\n\n\u003Cdetails>\n\u003Csummary>\u003Cstrong>DSPy\u003C\u002Fstrong>\u003C\u002Fsummary>\n\n```python\nimport dspy\nfrom agno.agents.dspy import DSPyAgent\nfrom agno.db.sqlite import SqliteDb\nfrom agno.os import AgentOS\n\ndspy.configure(lm=dspy.LM(\"openai\u002Fgpt-5.4\"))\n\nagent = DSPyAgent(\n    name=\"DSPy Assistant\",\n    program=dspy.ChainOfThought(\"question -> answer\"),\n)\n\nagent_os = AgentOS(agents=[agent], db=SqliteDb(db_file=\"agno.db\"), tracing=True)\napp = agent_os.get_app()\n```\n\n\u003C\u002Fdetails>\n\n### Run it\n\n```bash\nuv pip install -U 'agno[os]' openai\n\nexport OPENAI_API_KEY=sk-***\n\nfastapi dev workbench.py\n```\n\nIn ~20 lines, you get:\n\n- A FastAPI backend with 50+ endpoints\n- Streaming responses, persistent sessions, per-user isolation\n- Native OpenTelemetry tracing\n- Cron scheduling, human approval flows, and RBAC ready to enable\n\nAPI at `http:\u002F\u002Flocalhost:8000`. OpenAPI spec at `http:\u002F\u002Flocalhost:8000\u002Fdocs`.\n\n## Connect to the AgentOS UI\n\nThe [AgentOS UI](https:\u002F\u002Fos.agno.com) is your control plane. Use it to chat with your agents, inspect runs, view traces, manage sessions, and operate the system.\n\n1. Open [os.agno.com](https:\u002F\u002Fos.agno.com) and sign in.\n2. Click **\"Connect OS\"**\n3. Select **\"Local\"** to connect to a local AgentOS.\n4. Enter your endpoint URL (default: `http:\u002F\u002Flocalhost:8000`).\n5. Name it \"Local AgentOS\" and click **\"Connect\"**.\n\nOpen Chat, select your agent, and ask:\n\n> Tell me more about the project and the key files\n\nThe agent reads your workspace and answers grounded in what it actually finds. Try a follow-up like \"create a NOTES.md with three key takeaways\". The run pauses for your approval before the file is written, since `write_file` is a confirm-required tool by default.\n\nhttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Fadb38f55-1d9d-463e-8ca9-966bb6bdc37a\n\n## What AgentOS gives you\n\n- [**Production API**](https:\u002F\u002Fdocs.agno.com\u002Fruntime\u002Fserve-as-api). 50+ endpoints with SSE and websockets to build your product on.\n- [**Storage**](https:\u002F\u002Fdocs.agno.com\u002Fruntime\u002Fstorage). Sessions, memory, knowledge, and traces in your own database.\n- [**Context**](https:\u002F\u002Fdocs.agno.com\u002Fruntime\u002Fcontext). Live context across Slack, Drive, wikis, MCP, and custom sources.\n- [**Human approval**](https:\u002F\u002Fdocs.agno.com\u002Fruntime\u002Fhuman-approval). Pause runs for user confirmation, admin approval, or external execution.\n- [**Observability**](https:\u002F\u002Fdocs.agno.com\u002Fruntime\u002Fobservability). OpenTelemetry tracing, run history, and audit logs out of the box.\n- [**Security & auth**](https:\u002F\u002Fdocs.agno.com\u002Fruntime\u002Fsecurity-and-auth). JWT-based RBAC and multi-user, multi-tenant isolation.\n- [**Interfaces**](https:\u002F\u002Fdocs.agno.com\u002Fruntime\u002Finterfaces). Slack, Telegram, WhatsApp, Discord, AG-UI, A2A, or roll your own.\n- [**Scheduling**](https:\u002F\u002Fdocs.agno.com\u002Fruntime\u002Fscheduling). Cron-based scheduling and background jobs with no external infrastructure.\n- [**Deploy**](https:\u002F\u002Fdocs.agno.com\u002Fruntime\u002Fdeploy). Docker, Railway, AWS, GCP. Any container host works.\n\n## What you can build\n\nThree reference agents, all open source, all built on the same primitives:\n\n- [**Coda →**](https:\u002F\u002Fgithub.com\u002Fagno-agi\u002Fcoda) A Slack-native coding agent that ships PRs from your team chat.\n- [**Dash →**](https:\u002F\u002Fgithub.com\u002Fagno-agi\u002Fdash) A self-learning data agent grounded in six layers of context.\n- [**Scout →**](https:\u002F\u002Fgithub.com\u002Fagno-agi\u002Fscout) A self-learning context agent that manages enterprise knowledge.\n\n## Get started\n\n1. [Read the docs](https:\u002F\u002Fdocs.agno.com)\n2. [Build your first agent](https:\u002F\u002Fdocs.agno.com\u002Ffirst-agent)\n3. Explore the [cookbook](https:\u002F\u002Fgithub.com\u002Fagno-agi\u002Fagno\u002Ftree\u002Fmain\u002Fcookbook)\n\n## IDE integration\n\nAdd Agno docs as a source in your coding tools:\n\n**Cursor:** Settings → Indexing & Docs → Add `https:\u002F\u002Fdocs.agno.com\u002Fllms-full.txt`\n\nAlso works with VSCode, Windsurf, and similar tools.\n\n## Contributing\n\nSee the [contributing guide](https:\u002F\u002Fgithub.com\u002Fagno-agi\u002Fagno\u002Fblob\u002Fmain\u002FCONTRIBUTING.md).\n\n## Telemetry\n\nAgno logs which model providers are used to prioritize updates. Disable with `AGNO_TELEMETRY=false`.\n\n\u003Cp align=\"right\">\u003Ca href=\"#top\">↑ Back to top\u003C\u002Fa>\u003C\u002Fp>\n","Agno 是一个用于将代理作为生产软件运行的平台。它支持使用任何框架构建代理，并通过会话、跟踪、调度和基于角色的访问控制（RBAC）等功能将其作为生产服务运行。Agno 提供了一个SDK来构建具有记忆、知识和防护措施的代理及工作流，并集成了100多个工具；其运行时基于无状态、会话范围的FastAPI后端提供服务；并通过统一的控制平面实现系统的测试、监控与管理。此项目非常适合需要将AI代理快速部署到实际应用场景中的开发者，如自动化客服、智能助手等。",2,"2026-06-11 02:42:34","top_all"]