Up and running
in a minute.
Clone the source or use the one-line installer, add your LLM API key, and hand it the first task. GA grows its own environment through the agent — don't pre-install everything.
Two ways in.
Dependencies are deliberately tiered: the core needs only requests plus four lightweight packages for the browser bridge; the [ui] extra adds the bundled frontends. No Playwright, no LangChain, no browser binaries.
git clone https://github.com/lsdefine/GenericAgent.git && cd GenericAgent uv venv && uv pip install -e ".[ui]" cp mykey_template_en.py mykey.py # add your LLM API key python frontends/tui_v3.py # terminal UI
powershell -ExecutionPolicy Bypass -c "$env:GLOBAL=1; irm https://raw.githubusercontent.com/lsdefine/GenericAgent/main/assets/ga_install.ps1 | iex"
GLOBAL=1 bash -c "$(curl -fsSL https://raw.githubusercontent.com/lsdefine/GenericAgent/main/assets/ga_install.sh)"
Advanced capabilities are unlocked by instructing the agent — not by reading docs.
Each instruction makes GA read its pre-installed playbooks, install what is missing, adapt to your OS and persist the result into memory.
“Set up your web automation capability.” — GA guides you through the one manual step: dragging the bundled Chrome extension into chrome://extensions.
“Set up your OCR capability with rapidocr and save it to memory.”
“Set up your vision capability from the template in memory/.” — GA wires it to your existing keys and self-tests.
“Probe this system and set up your computer-use capability.”
python frontends/tui_v3.py
python launch.pyw
python frontends/tgapp.py · dcapp.py · fsapp.py
Install the APK from app.gaagent.ai
Everything else lives on GitHub.
Then give it a task it has never seen.
Watch it explore, succeed, and write down how — so it never has to figure it out again.