[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-72497":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":16,"stars7d":14,"stars30d":14,"stars90d":16,"forks30d":16,"starsTrendScore":16,"compositeScore":17,"rankGlobal":10,"rankLanguage":10,"license":18,"archived":19,"fork":20,"defaultBranch":21,"hasWiki":19,"hasPages":20,"topics":22,"createdAt":10,"pushedAt":10,"updatedAt":30,"readmeContent":31,"aiSummary":32,"trendingCount":16,"starSnapshotCount":16,"syncStatus":33,"lastSyncTime":34,"discoverSource":35},72497,"index","lmnr-ai\u002Findex","lmnr-ai","The SOTA Open-Source Browser Agent for autonomously performing complex tasks on the web","",null,"Python",2348,125,1,5,0,28.3,"Apache License 2.0",true,false,"main",[23,24,25,26,27,28,29],"ai","ai-agent","browser-agent","claude-3-7-sonnet","gemini-pro","llm","sota","2026-06-12 02:03:04","\u003Ca href=\"https:\u002F\u002Fgithub.com\u002Flmnr-ai\u002Findex\">![GitHub stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Flmnr-ai\u002Findex?style=social)\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fwww.ycombinator.com\u002Fcompanies\u002Flaminar-ai\">![Static Badge](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FY%20Combinator-S24-orange)\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fx.com\u002Flmnrai\">![X (formerly Twitter) Follow](https:\u002F\u002Fimg.shields.io\u002Ftwitter\u002Ffollow\u002Flmnrai)\u003C\u002Fa>\n\u003Ca href=\"https:\u002F\u002Fdiscord.gg\u002FnNFUUDAKub\"> ![Static Badge](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FJoin_Discord-464646?&logo=discord&logoColor=5865F2) \u003C\u002Fa>\n\n\u003Cpicture>\n  \u003Csource media=\"(prefers-color-scheme: dark)\" srcset=\".\u002Fstatic\u002Flogo_dark.png\">\n  \u003Csource media=\"(prefers-color-scheme: light)\" srcset=\".\u002Fstatic\u002Flogo_light.png\">\n  \u003Cimg alt=\"Laminar logo\" src=\".\u002Fstatic\u002Flogo_light.png\">\n\u003C\u002Fpicture>\n\n# Index\n\nIndex is a state-of-the-art open-source browser agent that autonomously executes complex web tasks. It turns any website into an accessible API and can be seamlessly integrated with just a few lines of code.\n\n- [x] Powered by reasoning LLMs with vision capabilities.\n    - [x] Gemini 2.5 Pro (really fast and accurate)\n    - [x] Claude 3.7 Sonnet with extended thinking (reliable and accurate)\n    - [x] OpenAI o4-mini (depending on the reasoning effort, provides good balance between speed, cost and accuracy)\n    - [x] Gemini 2.5 Flash (really fast, cheap, and good for less complex tasks)\n- [x] `pip install lmnr-index` and use it in your project\n- [x] `index run` to run the agent in the interactive CLI\n- [x] Supports structured output via Pydantic schemas for reliable data extraction.\n- [x] Index is also available as a [serverless API.](https:\u002F\u002Fdocs.lmnr.ai\u002Findex-agent\u002Fapi\u002Fgetting-started)\n- [x] You can also try out Index via [Chat UI](https:\u002F\u002Flmnr.ai\u002Fchat).\n- [x] Supports advanced [browser agent observability](https:\u002F\u002Fdocs.lmnr.ai\u002Findex-agent\u002Ftracing) powered by open-source platform [Laminar](https:\u002F\u002Fgithub.com\u002Flmnr-ai\u002Flmnr).\n\nprompt: go to ycombinator.com. summarize first 3 companies in the W25 batch and make new spreadsheet in google sheets.\n\nhttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F2b46ee20-81b6-4188-92fb-4d97fe0b3d6a\n\n## Documentation\n\nCheck out full documentation [here](https:\u002F\u002Fdocs.lmnr.ai\u002Findex-agent\u002Fgetting-started)\n\n## Quickstart\n\n### Install dependencies\n```bash\npip install lmnr-index 'lmnr[all]'\n\n# Install playwright\nplaywright install chromium\n```\n\n### Setup model API keys\n\nSetup your model API keys in `.env` file in your project root:\n```\nGEMINI_API_KEY=\nANTHROPIC_API_KEY=\nOPENAI_API_KEY=\n# Optional, to trace the agent's actions and record browser session\nLMNR_PROJECT_API_KEY=\n```\n\n### Run Index with code\n```python\nimport asyncio\nfrom index import Agent, GeminiProvider\nfrom pydantic import BaseModel\nfrom lmnr import Laminar\nimport os\n\n# to trace the agent's actions and record browser session\nLaminar.initialize()\n\n# Define Pydantic schema for structured output\nclass NewsSummary(BaseModel):\n    title: str\n    summary: str\n\nasync def main():\n\n    llm = GeminiProvider(model=\"gemini-2.5-pro-preview-05-06\")\n    agent = Agent(llm=llm)\n\n    # Example of getting structured output\n    output = await agent.run(\n        prompt=\"Navigate to news.ycombinator.com, find a post about AI, extract its title and provide a concise summary.\",\n        output_model=NewsSummary\n    )\n    \n    summary = NewsSummary.model_validate(output.result.content)\n    print(f\"Title: {summary.title}\")\n    print(f\"Summary: {summary.summary}\")\n    \nif __name__ == \"__main__\":\n    asyncio.run(main())\n```\n\n### Run Index with CLI\n\nIndex CLI features:\n- Browser state persistence between sessions\n- Follow-up messages with support for \"give human control\" action\n- Real-time streaming updates\n- Beautiful terminal UI using Textual\n\nYou can run Index CLI with the following command.\n```bash\nindex run\n```\n\nOutput will look like this:\n\n```\nLoaded existing browser state\n╭───────────────────── Interactive Mode ─────────────────────╮\n│ Index Browser Agent Interactive Mode                       │\n│ Type your message and press Enter. The agent will respond. │\n│ Press Ctrl+C to exit.                                      │\n╰────────────────────────────────────────────────────────────╯\n\nChoose an LLM model:\n1. Gemini 2.5 Flash\n2. Claude 3.7 Sonnet\n3. OpenAI o4-mini\nSelect model [1\u002F2] (1): 3\nUsing OpenAI model: o4-mini\nLoaded existing browser state\n\nYour message: go to lmnr.ai, summarize pricing page\n\nAgent is working...\nStep 1: Opening lmnr.ai\nStep 2: Opening Pricing page\nStep 3: Scrolling for more pricing details\nStep 4: Scrolling back up to view pricing tiers\nStep 5: Provided concise summary of the three pricing tiers\n```\n\n### Running CLI with a personal Chrome instance\n\nYou can use Index with personal Chrome browser instance instead of launching a new browser. Main advantage is that all your existing logged-in sessions will be available.\n\n```bash\n# Basic usage with default Chrome path\nindex run --local-chrome\n```\n\n## Use Index via API\n\nThe easiest way to use Index in production is with [serverless API](https:\u002F\u002Fdocs.lmnr.ai\u002Findex-agent\u002Fapi\u002Fgetting-started). Index API manages remote browser sessions, agent infrastructure and [browser observability](https:\u002F\u002Fdocs.lmnr.ai\u002Findex-agent\u002Fapi\u002Ftracing). To get started, create a project API key in [Laminar](https:\u002F\u002Flmnr.ai).\n\n### Install Laminar\n```bash\npip install lmnr\n```\n\n### Use Index via API\n```python\nfrom lmnr import Laminar, LaminarClient\n# you can also set LMNR_PROJECT_API_KEY environment variable\n\n# Initialize tracing\nLaminar.initialize(project_api_key=\"your_api_key\")\n\n# Initialize the client\nclient = LaminarClient(project_api_key=\"your_api_key\")\n\nfor chunk in client.agent.run(\n    stream=True,\n    model_provider=\"gemini\",\n    model=\"gemini-2.5-pro-preview-05-06\",\n    prompt=\"Navigate to news.ycombinator.com, find a post about AI, and summarize it\"\n):\n    print(chunk)\n    \n```\n\n\n## Browser agent observability\n\nBoth code run and API run provide advanced browser observability. To trace Index agent's actions and record browser session you simply need to initialize Laminar tracing before running the agent.\n\n```python\nfrom lmnr import Laminar\n\nLaminar.initialize(project_api_key=\"your_api_key\")\n```\n\nThen you will get full observability on the agent's actions synced with the browser session in the Laminar platform. Learn more about browser agent observability in the [documentation](https:\u002F\u002Fdocs.lmnr.ai\u002Findex-agent\u002Ftracing).\n\n\u003Cpicture>\n    \u003Cimg src=\".\u002Fstatic\u002Ftraces.png\" alt=\"Index observability\" width=\"800\"\u002F>\n\u003C\u002Fpicture>\n\n---\n\nMade with ❤️ by the [Laminar team](https:\u002F\u002Flmnr.ai)\n","Index 是一个先进的开源浏览器代理，能够自主执行复杂的网页任务。它通过具备视觉能力的推理大语言模型（如Gemini 2.5 Pro、Claude 3.7 Sonnet等）驱动，可以将任何网站转化为可访问的API，并且只需几行代码即可无缝集成到项目中。支持结构化输出和高级浏览器代理可观测性，适用于需要自动化处理复杂网络操作的场景，比如数据抓取、信息汇总以及自动化测试等。此外，Index还提供了服务器端API和聊天界面，方便用户以多种方式使用。",2,"2026-06-11 03:42:19","high_star"]