AgenticSeek

// Overview

AgenticSeek is an open-source, voice-enabled AI assistant built to run entirely on your own machine. It positions itself as a private, local alternative to Manus AI: it can browse the web, write and run code, and break large tasks into steps without sending your files, conversations, or searches to any cloud service.

The project is designed around local reasoning models such as Deepseek, Magistral, or Qwen running through providers like Ollama or LM Studio, so you can keep everything on your hardware. If your machine is not powerful enough, it can also connect to API providers such as OpenAI, Google Gemini, Deepseek, or TogetherAI instead.

AgenticSeek picks the best agent for each request automatically. You describe what you want, and it routes the job to the right specialist agent, whether that means searching and reading the web, writing code, or planning and coordinating a multi-step task across several agents.

// What it does

• Fully local and private: everything runs on your machine with no cloud dependency or data sharing
• Autonomous web browsing that can search, read pages, extract information, and fill in web forms hands-free
• Autonomous coding assistant that can write, debug, and run programs in Python, C, Go, Java, and more
• Smart agent selection that automatically chooses the best agent for each task you ask
• Plans and executes complex tasks by splitting big goals into steps across multiple AI agents
• Voice-enabled interaction with text-to-speech and speech-to-text (in progress)

// Getting started

AgenticSeek runs fully in Docker by default and ships with a web interface. You need Git, Python 3.10.x, and Docker Engine with Docker Compose installed before you start.

git clone https://github.com/Fosowl/agenticSeek.git
cd agenticSeek
mv .env.example .env

[MAIN]
is_local = True
provider_name = ollama
provider_model = deepseek-r1:14b
provider_server_address = 127.0.0.1:11434

export OLLAMA_HOST=0.0.0.0:11434
ollama serve

./start_services.sh full  # macOS / Linux
start start_services.cmd full  # Windows
# then open http://localhost:3000/

Commands and code are distilled from the project's own documentation — always check the official repo for the latest.

// When to use it

• Run a private AI assistant that browses, codes, and plans entirely offline so sensitive files and conversations never leave your machine
• Automate web research tasks such as searching, reading pages, and extracting or filling in information hands-free
• Generate, debug, and run code in languages like Python, C, Go, and Java without supervision
• Break a large multi-step project into smaller tasks and let multiple coordinated agents work through them

// How AgenticSeek compares

AgenticSeek alongside other open-source assistants & chatbots tools AI/TLDR tracks, ranked by GitHub stars.