AgenticSeek

What is AgenticSeek?

AgenticSeek is an open-source, voice-capable AI assistant that is built to run on your own hardware with local reasoning models. It browses the web by itself to search, read pages, extract information and fill in forms, and it writes, debugs and runs programs in languages such as Python, C, Go and Java. A routing layer picks the right agent for each request, and larger jobs are split into steps handled by several agents. The web interface runs in Docker alongside SearXNG for search, and there is also a command-line mode. You connect a local provider such as Ollama or LM Studio, or use an API key for OpenAI, Google Gemini, Deepseek, Hugging Face, TogetherAI, OpenRouter or MiniMax. API keys are optional when you run models locally. The project is free, and the authors say it started as a side project with no funding or roadmap.

Use cases

• Browse the web hands-free
• Write and debug code locally
• Plan multi-step projects
• Run an assistant on local models
• Extract information from web pages
• Organize files in a work folder

Who it's for

• Privacy-conscious users
• Developers
• Local LLM enthusiasts

Why use AgenticSeek

• Files, conversations and searches stay on your device when you use a local model, which suits private work.
• There is no subscription; the cost is the hardware you already have, plus any optional API usage.
• One assistant covers browsing, coding and planning, so you do not switch tools for each job.
• Automatic agent routing saves you from choosing a mode for every request.

AgenticSeek features

• Runs on local hardware with Ollama or LM Studio
• Autonomous web browsing with form filling
• Writes, debugs and runs code in several languages
• Automatic selection of the right agent per task
• Splits complex tasks into steps across agents
• Web interface in Docker and a CLI mode
• Optional cloud providers through API keys
• File upload in the web UI up to 50 MB per file
• Voice input and speech output (in progress)

AgenticSeek integrations

• Ollama
• LM Studio
• OpenAI
• Google Gemini
• Deepseek
• Hugging Face
• TogetherAI
• OpenRouter
• MiniMax
• SearXNG
• Docker

How to get started with AgenticSeek

1. Clone the repository, copy .env.example to .env and set WORK_DIR to the folder the agent may use.
2. Start Ollama, LM Studio or another provider, then set provider_name and provider_model in config.ini.
3. Run ./start_services.sh full to launch the Docker services, or install the package and use uv run cli.py.
4. Open localhost:3000, paste the API token from the backend log, and ask for a search, a program or a plan.
5. Say or type goodbye to exit when you finish.

AgenticSeek limitations

• Running local models needs a GPU able to run a 14B-class model such as Magistral, Qwen or Deepseek
• Voice features are still in progress
• First Docker start downloads images and can take up to 30 minutes
• Prompts are tuned for Deepseek-style models, so GPT-4o and Gemini can behave less reliably

AgenticSeek pricing

AgenticSeek is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

AgenticSeek alternatives

The main AgenticSeek alternatives listed on Debuly are OpenManus, Manus, OpenHands, AutoGPT, Open Interpreter and Evaligo. The table compares pricing and platforms; order is A–Z.

1. 

   AutoGPT

   Agent platform with a plain-English AutoPilot, a visual workflow builder and self-hosting, plus a $50/month Pro plan with a 7-day trial.

   Automation and agentsDeveloper platformsProductivity
   FreemiumVisit AutoGPT (official site)
2. 

   Evaligo

   Visual AI workflow builder with prompt testing and evaluation, deploying flows as REST APIs, with a free plan and a $7/month Pro plan.

   Automation and agentsDeveloper platformsCoding
   FreemiumVisit Evaligo (official site)
3. 

   Manus

   General AI agent that carries out multi-step tasks such as research, slides, websites and design from a single prompt.

   Automation and agentsChat assistantsProductivity
   FreemiumVisit Manus (official site)
4. 

   Open Interpreter

   Open-source terminal coding agent, built in Rust on Codex, that emulates the agent harness suited to each low-cost model.

   CodingAutomation and agentsDeveloper platforms
   Open sourceVisit Open Interpreter (official site)
5. 

   OpenHands

   Open-source AI coding agent platform with a free Agent Canvas, a Teams plan, and self-hosted Enterprise control

   CodingAutomation and agentsDeveloper platforms
   FreemiumVisit OpenHands (official site)
6. 

   OpenManus

   Open-source framework for building general AI agents that you run yourself, with browser automation and MCP tool support

   Automation and agentsDeveloper platforms
   Open sourceVisit OpenManus (official site)

AgenticSeek FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: none collected yet, so a monogram is shown.
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.