agenticSeek - Moses Yebei

Abstract

Fully Local Manus AI. No APIs, No $200 monthly bills. Enjoy an autonomous agent that thinks, browses the web, and code for the sole cost of electricity. 🔔 Official updates only via twitter @Martin993886460 (Beware of fake account)

The Problem

AI agents are cool and all, but who has the cash to blow $200 a month on API calls? Beyond the subscription fees, there's the even bigger issue of privacy. Do you really want your personal data and queries sent off to some random server farm? No, thanks. Enter agenticSeek, the self-contained, fully local AI assistant that keeps your data where it belongs—on your machine.

What This Does

agenticSeek is basically your personal Jarvis, minus the billionaire tech mogul backstory. It runs entirely on your hardware, using Python, Flask, React, and Docker to spin up a suite of modular AI agents. You’ve got file agents in sources/agents/fileagent.py, coding helpers like sources/tools/PyInterpreter.py, and even a browser agent (sources/agents/browseragent.py) for autonomous web searches.

The directory structure is clean and logical. For example, anything to do with large language models lives in llmserver/. Configs like .env.example, docker-compose.yml, and various Dockerfile setups keep deployment straightforward. Speaking of, the project is very Docker-friendly—Dockerfiles for the backend, frontend, and LLM server are all included. You’ll appreciate that if you’ve ever spent hours debugging dependencies.

One standout feature is how it picks the right agent for the job. Ask it to analyze a .zip of resumes? It knows to fetch the fileagent. Want to scrape data from a website? That’s the browseragent’s turf.

Real-World Use

Imagine this: You’re working on a side project and need to write a Python script to scrape product data from multiple sites. Instead of googling for hours and writing boilerplate code, you boot up agenticSeek. Tell it what you need done, and it’ll spin up the browseragent, fetch the necessary data, and even write a script in sources/tools/webSearch.py to automate the task. Oh, and all of this happens locally—your data, your rules.

Setup is simple. Clone the repo, install Python 3.10.x, fire up Docker with docker-compose up, and you’re good to go. The cli.py script is your main entry point if you’re more of a terminal junkie.

The Bottom Line

agenticSeek is a no-bullshit, privacy-first AI assistant for power users. It’s loaded with functionality and doesn’t shove recurring fees or cloud nonsense down your throat. That said, it’s not for the faint of heart—there’s no hand-holding here, and you’ll need to know your way around Docker and Python. If that’s you, though, this could be your new favorite tool.

What the analyser found

Deployment readiness

3/7

Container image
CI pipeline
Lockfile committed
Test suite
README
License
No committed secrets

Composition

137 files

Python46
Markdown12
JavaScript12
Shell7
YAML5
JSON5

ReactFlaskExpressDocker

Architecture

Top-level areas of the codebase, sized by module count. Arrows show how many imports cross from one area into another.

graph TD g0["sources 35 modules"] g1["frontend 9 modules"] g2["llm_server 6 modules"] g3["tests 5 modules"] g4["(root) 3 modules"] g4 -->|12| g0 g3 -->|5| g0

Module dependencies

The 10 most depended-upon modules of 58, from static import analysis. Red outline marks a module in an import cycle.

graph LR n0_sources_utility["utility"] n1_sources_tools_tools["tools"] n2_sources_logger["logger"] n3_sources_agents_agent["agent"] n4_sources_memory["memory"] n5_sources_agents_browser_agent["browser_agent"] n6_ources_tools_BashInterpreter["BashInterpreter"] n7_sources_tools_fileFinder["fileFinder"] n8_sources_agents_code_agent["code_agent"] n9_sources_agents_casual_agent["casual_agent"] n3_sources_agents_agent --> n4_sources_memory n3_sources_agents_agent --> n0_sources_utility n5_sources_agents_browser_agent --> n0_sources_utility n5_sources_agents_browser_agent --> n3_sources_agents_agent n5_sources_agents_browser_agent --> n2_sources_logger n5_sources_agents_browser_agent --> n4_sources_memory n9_sources_agents_casual_agent --> n0_sources_utility n9_sources_agents_casual_agent --> n3_sources_agents_agent n9_sources_agents_casual_agent --> n7_sources_tools_fileFinder n9_sources_agents_casual_agent --> n6_ources_tools_BashInterpreter n9_sources_agents_casual_agent --> n4_sources_memory n8_sources_agents_code_agent --> n0_sources_utility n8_sources_agents_code_agent --> n3_sources_agents_agent n8_sources_agents_code_agent --> n6_ources_tools_BashInterpreter n8_sources_agents_code_agent --> n7_sources_tools_fileFinder n8_sources_agents_code_agent --> n2_sources_logger n8_sources_agents_code_agent --> n4_sources_memory n4_sources_memory --> n0_sources_utility