LLM Endpoint
Set your model provider values and API keys for chat and tool-enabled runs.
- OPENAI_API_KEY
- OPENAI_BASE_URL
- MODEL_NAME
Get Started
Install CATBot and run your first end-to-end task.
This quick path takes you from prerequisites to a working request loop with environment setup, runtime launch, and first-task validation in the open-source CATBot project.
Prerequisites
Baseline environment checklist.
Python 3.10+
Required for backend services and tool integrations.
Node.js 18+
Required for frontend dependencies and runtime scripts.
Modern browser
Responsive HTML client works across desktop and iOS Safari.
Git + terminal
Used for cloning and launching project commands.
Optional: Docker
Needed for certain AutoGen code-execution workflows.
macOS / Linux
git clone <catbot-repo>
cd CATBot
./install.sh
cp .env.example .env
python src/servers/proxy_server.py
Windows PowerShell
git clone <catbot-repo>
cd CATBot
./install.ps1
Copy-Item .env.example .env
python src/servers/proxy_server.py
Configuration
First values to set in .env.
Feature Toggles
Enable optional channels and tools as needed for your use case.
- TELEGRAM_TOOLS_ENABLED
- CODEX_ENABLED
- TASK_EXECUTION_MAX_ITERATIONS
First task prompt
Start with a small scoped change and ask for validation. Example: "Update the landing page copy and include file references in the final summary."
After first run
- Review output quality and adjust system prompt guidance.
- Enable only the tools your team actually needs.
- Document repeatable workflows in project docs.
- Move to contribution flow for custom skills and modules.
Launch
You now have the setup path to run CATBot in production-style workflows.
Use the community and architecture guides to scale from solo usage to team adoption.