fosowl/agenticseek

📈 Trending★ 26,369PythonAudience · vibe coderComplexity · 4/5ActiveLicenseSetup · hard

Mindmap

mindmap
  root((AgenticSeek))
    What it does
      Autonomous web browsing
      Code execution
      Multi-step task completion
    Key features
      Fully local and private
      No cloud dependency
      No API costs
    Tech stack
      Local AI models
      Docker
      Python
    Use cases
      Research automation
      Coding assistance
      Data gathering
    Trade-offs
      High hardware needs
      GPU memory required
      Setup complexity

mindmap root((AgenticSeek)) What it does Autonomous web browsing Code execution Multi-step task completion Key features Fully local and private No cloud dependency No API costs Tech stack Local AI models Docker Python Use cases Research automation Coding assistance Data gathering Trade-offs High hardware needs GPU memory required Setup complexity

Things people build with this

USE CASE 1

Build an autonomous research assistant that finds competitors, compares pricing, and generates reports without manual web browsing.

USE CASE 2

Create a coding helper that writes, tests, and debugs code across multiple files based on natural language descriptions.

USE CASE 3

Automate data gathering tasks like scraping websites, extracting information, and organizing results into spreadsheets.

Tech stack

PythonDockerOllamaLM Studio

Getting it running

Difficulty · hard Time to first run · 1day+

Requires Docker, local LLM setup (Ollama or LM Studio), and significant resource allocation for autonomous agent execution.

Use it freely, but any project you distribute that includes this code must also be GPL-licensed and open source.

In plain English

AgenticSeek is a fully local, privacy-first AI assistant that can autonomously browse the web, write and run code, and handle complex multi-step tasks, all without sending your data to any cloud service. It's positioned as a free alternative to Manus AI, which is a similar agentic AI product that costs money and runs on external servers. The key distinction is that everything runs on your own computer using locally installed AI models (via tools like Ollama or LM Studio). Your conversations, files, and web searches stay entirely on your device. There are no monthly API bills as long as you have the hardware to run the AI models locally. For a vibe coder or founder, this is the kind of tool that could serve as an autonomous research assistant or coding helper. You describe a task in natural language, say, "find the top 5 competitors in the project management space, compare their pricing, and create a spreadsheet", and AgenticSeek figures out how to accomplish it step by step: searching the web, reading pages, writing code, and producing results. The trade-off is significant hardware requirements. Running the recommended AI models locally needs a modern GPU with enough memory (typically 16GB+ VRAM for capable models). Setup also requires comfort with Docker and some technical configuration. It's currently an active work-in-progress side project that grew unexpectedly, so rough edges should be expected. For someone with a capable Mac with Apple silicon or a gaming PC with a good GPU who wants to experiment with autonomous AI agents without cloud costs or privacy concerns, this is worth exploring.

Copy-paste prompts

Prompt 1

Set up AgenticSeek locally on my machine with Docker and configure it to use Ollama for running AI models.

Prompt 2

Use AgenticSeek to research the top 5 project management tools, compare their pricing plans, and create a comparison spreadsheet.

Prompt 3

Build a custom workflow in AgenticSeek that autonomously searches for information, writes Python scripts, and executes them to solve a specific problem.

Prompt 4

Configure AgenticSeek to act as a coding assistant that can read my codebase, suggest improvements, and implement changes across multiple files.

Open on GitHub → Explain another repo

Generated 2026-05-18 · Model: sonnet-4-6 · Verify against the repo before relying on details.