{"id":26896912,"url":"https://github.com/Fosowl/agenticSeek","last_synced_at":"2025-04-01T04:02:27.613Z","repository":{"id":278588190,"uuid":"935604638","full_name":"Fosowl/agenticSeek","owner":"Fosowl","description":"A open, local Manus AI alternative. Powered with Deepseek R1. No APIs, no $456 monthly bills. Enjoy an AI agent that reason, code, and browse with no worries.","archived":false,"fork":false,"pushed_at":"2025-03-29T11:29:44.000Z","size":17993,"stargazers_count":688,"open_issues_count":10,"forks_count":89,"subscribers_count":7,"default_branch":"main","last_synced_at":"2025-03-29T11:31:47.500Z","etag":null,"topics":["agentic-ai","agents","ai","autonomous-agents","deepseek","deepseek-r1","llm","llm-agents","voice-assistant"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Fosowl.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2025-02-19T18:03:06.000Z","updated_at":"2025-03-29T10:49:18.000Z","dependencies_parsed_at":"2025-03-07T10:26:44.325Z","dependency_job_id":"6f5a6004-82c1-4f63-8481-f5fd04d2f769","html_url":"https://github.com/Fosowl/agenticSeek","commit_stats":null,"previous_names":["fosowl/agenticseek"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Fosowl%2FagenticSeek","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Fosowl%2FagenticSeek/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Fosowl%2FagenticSeek/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Fosowl%2FagenticSeek/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Fosowl","download_url":"https://codeload.github.com/Fosowl/agenticSeek/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":246580468,"owners_count":20800111,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agentic-ai","agents","ai","autonomous-agents","deepseek","deepseek-r1","llm","llm-agents","voice-assistant"],"created_at":"2025-04-01T04:02:23.083Z","updated_at":"2025-04-01T04:02:27.607Z","avatar_url":"https://github.com/Fosowl.png","language":"Python","funding_links":[],"categories":["AI Agent Frameworks","Python","A01_文本生成_文本对话","Architecture \u0026 Workflow \u003ca id=\"architecture--workflow\"\u003e\u003c/a\u003e","HarmonyOS","Repos","voice-assistant","Other AI Agents","Table of Contents","🧠 AI Applications \u0026 Platforms","🚀 Specialized Agents","Autonomous Research \u0026 Content Generation","App","🤖 AI \u0026 Machine Learning"],"sub_categories":["General Agent Frameworks","大语言对话模型及数据","Dynamic Single‑Agent Workflow \u003ca id=\"dynamic-single-agent-workflow\"\u003e\u003c/a\u003e","Windows Manager","General-Purpose Agent Frameworks","Agent","Tools","💻 Coding Agents","Programming Languages"],"readme":"\n# AgenticSeek: Manus-like AI powered by Deepseek R1 Agents.\n\n\n**A fully local alternative to Manus AI**, a voice-enabled AI assistant that codes, explores your filesystem, browse the web and correct it's mistakes all without sending a byte of data to the cloud. Built with reasoning models like DeepSeek R1, this autonomous agent runs entirely on your hardware, keeping your data private.\n\n[![Visit AgenticSeek](https://img.shields.io/static/v1?label=Website\u0026message=AgenticSeek\u0026color=blue\u0026style=flat-square)](https://fosowl.github.io/agenticSeek.html) ![License](https://img.shields.io/badge/license-GPL--3.0-green) [![Discord](https://img.shields.io/badge/Discord-Join%20Us-7289DA?logo=discord\u0026logoColor=white)](https://discord.gg/4Ub2D6Fj)\n\n\u003e 🛠️ **Work in Progress** – Looking for contributors!\n\n![alt text](./media/whale_readme.jpg)\n\n\u003e *Do a deep search of AI startup in Osaka and Tokyo, find at least 5, then save in the research_japan.txt file*\n\n\u003e *Can you make a tetris game in C ?*\n\n\u003e *I would like to setup a new project file index as mark2.*\n\n\n### agenticSeek can now plan tasks!\n\n![alt text](./media/exemples/demo_image.png)\n\n*See media/examples for other use case screenshots.*\n\n## Features:\n\n- **100% Local**: No cloud, runs on your hardware. Your data stays yours.\n\n- **Filesystem interaction**: Use bash to navigate and manipulate your files effortlessly.\n\n- **Autonomous Coding**: Can write, debug, and run code in Python, C, Golang and more languages on the way.\n\n- **Agent routing**: Automatically picks the right agent for the job.\n\n- **Planning**: For complex tasks, spins up multiple agents to plan and execute.\n\n- **Autonomous Web Browsing**: Autonomous web navigation.\n\n- **Memory**: Efficient memory and sessions management. \n\n---\n\n## **Installation**\n\nMake sure you have chrome driver, docker and python3.10 (or newer) installed.\n\nFor issues related to chrome driver, see the **Chromedriver** section.\n\n### 1️⃣ **Clone the repository and setup**\n\n```sh\ngit clone https://github.com/Fosowl/agenticSeek.git\ncd agenticSeek\nmv .env.example .env\n```\n\n### 2️ **Create a virtual env**\n\n```sh\npython3 -m venv agentic_seek_env\nsource agentic_seek_env/bin/activate     \n# On Windows: agentic_seek_env\\Scripts\\activate\n```\n\n### 3️⃣ **Install package**\n\n**Automatic Installation:**\n\n```sh\n./install.sh\n```\n\n**Manually:**\n\n```sh\npip3 install -r requirements.txt\n# or\npython3 setup.py install\n```\n\n\n## Run locally on your machine\n\n**We recommend using at least Deepseek 14B, smaller models struggle with tool use and forget quickly the context.**\n\n### 1️⃣ **Download Models**  \n\nMake sure you have [Ollama](https://ollama.com/) installed.\n\nDownload the `deepseek-r1:14b` model from [DeepSeek](https://deepseek.com/models)\n\n```sh\nollama pull deepseek-r1:14b\n```\n\n### 2️ **Run the Assistant (Ollama)**  \n\nStart the ollama server\n```sh\nollama serve\n```\n\nChange the config.ini file to set the provider_name to `ollama` and provider_model to `deepseek-r1:14b`\n\nNOTE: `deepseek-r1:14b`is an example, use a bigger model if your hardware allow it.\n\n```sh\n[MAIN]\nis_local = True\nprovider_name = ollama\nprovider_model = deepseek-r1:14b\nprovider_server_address = 127.0.0.1:11434\n```\n\nstart all services :\n\n```sh\nsudo ./start_services.sh\n```\n\nRun the assistant:\n\n```sh\npython3 main.py\n```\n\n*See the **Usage** section if you don't understand how to use it*\n\n*See the **Known issues** section if you are having issues*\n\n*See the **Run with an API** section if your hardware can't run deepseek locally*\n\n*See the **Config** section for detailled config file explanation.*\n\n---\n\n## Usage\n\nWarning: currently the system that choose the best AI agent routing system will work poorly with non-english text. This is because the agent routing currently use a model that was trained on english text. We are working hard to fix this. Please use english for now.\n\n\nMake sure the services are up and running with `./start_services.sh` and run the agenticSeek with `python3 main.py`\n\n```sh\nsudo ./start_services.sh\npython3 main.py\n```\n\nYou will be prompted with `\u003e\u003e\u003e `\nThis indicate agenticSeek await you type for instructions.\nYou can also use speech to text by setting `listen = True` in the config.\n\nTo exit, simply say `goodbye`.\n\nHere are some example usage:\n\n### Coding/Bash\n\n\u003e *Help me with matrix multiplication in Golang*\n\n\u003e *Scan my network with nmap, find if any suspicious devices is connected*\n\n\u003e *Make a snake game in python*\n\n### Web search\n\n\u003e *Do a web search to find cool tech startup in Japan working on cutting edge AI research*\n\n\u003e *Can you find on the internet who created agenticSeek?*\n\n\u003e *Can you find on which website I can buy a rtx 4090 for cheap*\n\n### File system\n\n\u003e *Hey can you find where is million_dollars_contract.pdf i lost it*\n\n\u003e *Show me how much space I have left on my disk*\n\n\u003e *Find and read the README.md and follow the install instruction*\n\n### Casual\n\n\u003e *Tell me about France*\n\n\u003e *What is the meaning of life ?*\n\n\u003e *Should I take creatine before or after workout?*\n\n\nAfter you type your query, agenticSeek will allocate the best agent for the task.\n\nBecause this is an early prototype, the agent routing system might not always allocate the right agent based on your query.\n\nTherefore, you should be very explicit in what you want and how the AI might proceed for example if you want it to conduct a web search, do not say:\n\n`Do you know some good countries for solo-travel?`\n\nInstead, ask:\n\n`Do a web search and find out which are the best country for solo-travel`\n\n---\n\n## **Run the LLM on your own server**  \n\nIf you have a powerful computer or a server that you can use, but you want to use it from your laptop you have the options to run the LLM on a remote server. \n\n### 1️⃣  **Set up and start the server scripts** \n\nOn your \"server\" that will run the AI model, get the ip address\n\n```sh\nip a | grep \"inet \" | grep -v 127.0.0.1 | awk '{print $2}' | cut -d/ -f1\n```\n\nNote: For Windows or macOS, use ipconfig or ifconfig respectively to find the IP address.\n\n**If you wish to use openai based provider follow the *Run with an API*  section.**\n\nClone the repository and enter the `server/`folder.\n\n\n```sh\ngit clone --depth 1 https://github.com/Fosowl/agenticSeek.git\ncd agenticSeek/server/\n```\n\nInstall server specific requirements:\n\n```sh\npip3 install -r requirements.txt\n```\n\nRun the server script.\n\n```sh\npython3 app.py --provider ollama --port 3333\n```\n\nYou have the choice between using `ollama` and `llamacpp` as a LLM service.\n\n### 2️⃣ **Run it** \n\nNow on your personal computer:\n\nChange the `config.ini` file to set the `provider_name` to `server` and `provider_model` to `deepseek-r1:14b`.\nSet the `provider_server_address` to the ip address of the machine that will run the model.\n\n```sh\n[MAIN]\nis_local = False\nprovider_name = server\nprovider_model = deepseek-r1:14b\nprovider_server_address = x.x.x.x:3333\n```\n\nRun the assistant:\n\n```sh\nsudo ./start_services.sh\npython3 main.py\n```\n\n## **Run with an API**  \n\nSet the desired provider in the `config.ini`\n\n```sh\n[MAIN]\nis_local = False\nprovider_name = openai\nprovider_model = gpt-4o\nprovider_server_address = 127.0.0.1:5000\n```\n\nWARNING: Make sure there is not trailing space in the config.\n\nSet `is_local` to True if using a local openai-based api.\n\nChange the IP address if your openai-based api run on your own server.\n\nRun the assistant:\n\n```sh\nsudo ./start_services.sh\npython3 main.py\n```\n\n---\n\n## Speech to Text\n\nThe speech-to-text functionality is disabled by default. To enable it, set the listen option to True in the config.ini file:\n\n```\nlisten = True\n```\n\nWhen enabled, the speech-to-text feature listens for a trigger keyword, which is the agent's name, before it begins processing your input. You can customize the agent's name by updating the `agent_name` value in the *config.ini* file:\n\n```\nagent_name = Friday\n```\n\nFor optimal recognition, we recommend using a common English name like \"John\" or \"Emma\" as the agent name\n\nOnce you see the transcript start to appear, say the agent's name aloud to wake it up (e.g., \"Friday\").\n\nSpeak your query clearly.\n\nEnd your request with a confirmation phrase to signal the system to proceed. Examples of confirmation phrases include:\n```\n\"do it\", \"go ahead\", \"execute\", \"run\", \"start\", \"thanks\", \"would ya\", \"please\", \"okay?\", \"proceed\", \"continue\", \"go on\", \"do that\", \"go it\", \"do you understand?\"\n```\n\n## Config\n\nExample config:\n```\n[MAIN]\nis_local = True\nprovider_name = ollama\nprovider_model = deepseek-r1:1.5b\nprovider_server_address = 127.0.0.1:11434\nagent_name = Friday\nrecover_last_session = False\nsave_session = False\nspeak = False\nlisten = False\nwork_dir =  /Users/mlg/Documents/ai_folder\njarvis_personality = False\n[BROWSER]\nheadless_browser = False\nstealth_mode = False\n```\n\n**Explanation**:\n\n- is_local -\u003e Runs the agent locally (True) or on a remote server (False).\n- provider_name -\u003e The provider to use (one of: `ollama`, `server`, `lm-studio`, `deepseek-api`)\n- provider_model -\u003e The model used, e.g., deepseek-r1:1.5b.\n- provider_server_address -\u003e Server address, e.g., 127.0.0.1:11434 for local. Set to anything for non-local API.\n- agent_name -\u003e Name of the agent, e.g., Friday. Used as a trigger word for TTS.\n- recover_last_session -\u003e Restarts from last session (True) or not (False).\n- save_session -\u003e Saves session data (True) or not (False).\n- speak -\u003e Enables voice output (True) or not (False).\n- listen -\u003e listen to voice input (True) or not (False).\n- work_dir -\u003e Folder the AI will have access to. eg: /Users/user/Documents/.\n- jarvis_personality -\u003e Uses a JARVIS-like personality (True) or not (False). This simply change the prompt file.\n- headless_browser -\u003e Runs browser without a visible window (True) or not (False).\n- stealth_mode -\u003e Make bot detector time harder. Only downside is you have to manually install the anticaptcha extension.\n\n## Providers\n\nThe table below show the available providers:\n\n| Provider  | Local? | Description                                               |\n|-----------|--------|-----------------------------------------------------------|\n| ollama    | Yes    | Run LLMs locally with ease using ollama as a LLM provider |\n| server    | Yes    | Host the model on another machine, run your local machine |\n| lm-studio  | Yes    | Run LLM locally with LM studio (set `provider_name` to `lm-studio`)|\n| openai    | No     | Use ChatGPT API (non-private)                             |\n| deepseek-api  | No     | Deepseek API (non-private)                                |\n| huggingface| No    | Hugging-Face API (non-private)                            |\n\nTo select a provider change the config.ini:\n\n```\nis_local = False\nprovider_name = openai\nprovider_model = gpt-4o\nprovider_server_address = 127.0.0.1:5000\n```\n`is_local`: should be True for any locally running LLM, otherwise False.\n\n`provider_name`: Select the provider to use by it's name, see the provider list above.\n\n`provider_model`: Set the model to use by the agent.\n\n`provider_server_address`: can be set to anything if you are not using the server provider.\n\n# Known issues\n\n## Chromedriver Issues\n\n**Known error #1:** *chromedriver mismatch*\n\n`Exception: Failed to initialize browser: Message: session not created: This version of ChromeDriver only supports Chrome version 113\nCurrent browser version is 134.0.6998.89 with binary path`\n\nThis happen if there is a mismatch between your browser and chromedriver version.\n\nYou need to navigate to download the latest version:\n\nhttps://developer.chrome.com/docs/chromedriver/downloads\n\nIf you're using Chrome version 115 or newer go to:\n\nhttps://googlechromelabs.github.io/chrome-for-testing/\n\nAnd download the chromedriver version matching your OS.\n\n![alt text](./media/chromedriver_readme.png)\n\nIf this section is incomplete please raise an issue.\n\n## FAQ\n\n**Q: What hardware do I need?**  \n\n7B Model: GPU with 8GB VRAM.\n14B Model: 12GB GPU (e.g., RTX 3060).\n32B Model: 24GB+ VRAM.\n\n**Q: Why Deepseek R1 over other models?**  \n\nDeepseek R1 excels at reasoning and tool use for its size. We think it’s a solid fit for our needs other models work fine, but Deepseek is our primary pick.\n\n**Q: I get an error running `main.py`. What do I do?**  \n\nEnsure Ollama is running (`ollama serve`), your `config.ini` matches your provider, and dependencies are installed. If none work feel free to raise an issue.\n\n**Q: Can it really run 100% locally?**  \n\nYes with Ollama or Server providers, all speech to text, LLM and text to speech model run locally. Non-local options (OpenAI or others API) are optional.\n\n**Q: How come it is older than manus ?**\n\nwe started this a fun side project to make a fully local, Jarvis-like AI. However, with the rise of Manus, we saw the opportunity to redirected some tasks to make yet another alternative.\n\n**Q: How is it better than manus  ?**\n\nIt's not but we prioritizes local execution and privacy over cloud based approach. It’s a fun, accessible alternative!\n\n## Contribute\n\nWe’re looking for developers to improve AgenticSeek! Check out open issues or discussion.\n\n[![Star History Chart](https://api.star-history.com/svg?repos=Fosowl/agenticSeek\u0026type=Date)](https://www.star-history.com/#Fosowl/agenticSeek\u0026Date)\n\n## Authors:\n \u003e [Fosowl](https://github.com/Fosowl)\n \u003e [steveh8758](https://github.com/steveh8758) \n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FFosowl%2FagenticSeek","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FFosowl%2FagenticSeek","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FFosowl%2FagenticSeek/lists"}