GenericAgent

What is GenericAgent?

GenericAgent is a small, self-evolving agent framework that lets an LLM operate a local computer. The core is about 3,000 lines of seed code with nine atomic tools and an agent loop of roughly 100 lines, covering code execution, file reading, writing and patching, web scanning, browser JavaScript control and asking the user for confirmation. A bundled browser driver works inside a real browser so existing logins are preserved, and the agent can also use the keyboard and mouse, screen vision, OCR and Android devices through ADB. Each time it solves a new task it saves the steps as a skill kept in a layered memory, so a repeated request becomes a one-line call. It works with Claude, Gemini, Kimi, MiniMax and other models and runs through a terminal UI, a Streamlit web UI, or Telegram, Discord and Lark bots. You clone the repository, add your own model API key and run it on Python 3.11 or 3.12. The project is free and MIT-licensed.

Use cases

• Automate tasks in a real browser
• Control a desktop with mouse and keyboard
• Build reusable skills from repeated tasks
• Run an agent from Telegram or Discord
• Script scheduled web summaries
• Drive Android devices through ADB

Who it's for

• Developers
• AI agent researchers
• Automation enthusiasts

Why use GenericAgent

• The small codebase is easy to read, audit and modify compared with large agent frameworks.
• Skills accumulate as you use it, so repeat tasks become short commands.
• Controlling a real browser keeps existing logins and avoids sandbox-only automation.
• It is model-agnostic, so you can switch between major LLM providers.
• It is free, with the only cost being the API usage of the model you connect.

GenericAgent features

• About 3K lines of core code and a roughly 100-line agent loop
• Nine atomic tools including code_run, file_read, file_write and web_execute_js
• Skills saved automatically after each solved task
• Layered memory from meta rules to session archives
• Real-browser control that keeps login sessions
• Screen vision, OCR and computer-use capabilities added on request
• Android control over ADB
• Terminal UI, Streamlit UI and Telegram, Discord and Lark bots
• Works with Claude, Gemini, Kimi, MiniMax and other models

GenericAgent integrations

• Claude
• Gemini
• Kimi
• MiniMax
• Telegram
• Discord
• Lark
• Chrome
• ADB
• Streamlit

How to get started with GenericAgent

1. Clone the repository and install it with uv, or use the one-line installer.
2. Copy the key template to mykey.py and add your LLM API key.
3. Launch the terminal UI or Streamlit app, or start a Telegram, Discord or Lark bot.
4. Tell the agent what to do; it perceives the environment, runs tools and writes the result to memory.
5. Ask it to set up extra abilities such as web automation, OCR or vision, and reuse saved skills next time.

GenericAgent limitations

• Gives the model system-level control of your computer, so run it with care
• Needs Python 3.11 or 3.12; Python 3.14 is incompatible
• Pre-installed SOPs are written in Chinese and were tuned on Windows
• You must supply your own LLM API key
• Token-efficiency claims are the project's own

GenericAgent pricing

GenericAgent 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.

GenericAgent alternatives

The main GenericAgent alternatives listed on Debuly are OpenClaw, Claude Code, Open Interpreter, AutoGPT, Browser Use and OpenHands. 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. 

   Browser Use

   Cloud browsers and hosted web agents for AI, plus an MIT-licensed open-source browser agent, billed pay-as-you-go from $0.02 per hour.

   Developer platformsAutomation and agentsCoding
   PaidVisit Browser Use (official site)
3. 

   Claude Code

   Anthropic's coding agent that fixes bugs, writes tests and runs migrations from your terminal, IDE, desktop app, web or Slack.

   CodingAutomation and agentsDeveloper platforms
   PaidVisit Claude Code (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. 

   OpenClaw

   Open-source personal AI assistant that runs on your own devices and takes actions for you through WhatsApp, Telegram, Slack, Discord and other chat apps.

   Automation and agentsChat assistantsProductivity
   Open sourceVisit OpenClaw (official site)
6. 

   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)

GenericAgent 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.