Get up and running with Agentao in under 2 minutes!
pip install agentao
echo "OPENAI_API_KEY=your-api-key-here" > .env
Or copy the example file if you have a source checkout:
cp .env.example .env
# Edit .env and add your API key
agentao
Once running, try these commands:
You: /help
Shows all available commands and features.
You: Read the file README.md
The agent will use the read_file tool to read and display the file.
You: Search for all Python files in this directory
The agent will use the glob tool to find all .py files.
You: /skills
Shows all available skills.
You: Remember that I prefer using uv for Python projects
Saves this preference to memory using the save_memory tool.
All commands start with /:
/help - Show help message/model - List or switch models
/model - Show current and available models/model gpt-4o - Switch to GPT-4o/model claude-sonnet-4-6 - Switch to Claude Sonnet/clear - Clear conversation history/status - Show conversation status (includes current model)/skills - List available skills/memory - Show saved memories/exit or /quit - Exit the programNote: Regular messages (without /) are sent to the AI agent.
You: Create a new file called test.py with a hello world function
You: Read the file test.py
You: Replace the hello function with a greeting function that takes a name
You: Search for Python async best practices
You: Fetch the content from the first result
You: Activate the pdf skill
You: Now help me merge multiple PDFs
You: /model (show current model and available models)
You: /model gpt-4o (switch to GPT-4o)
You: /skills (list all skills)
You: /memory (show memories)
You: /status (show status and current model)
You: /clear (clear history)
You: Find all TODO comments in Python files
You: Search for function definitions in the utils module
You: Use the codebase investigator to analyze the project structure
Be specific: The more specific your request, the better the agent can help you.
Use tools naturally: Just ask in natural language - the agent will figure out which tools to use.
Check memory: Use the /memory command to see what the agent has remembered about your preferences.
Activate skills: For specialized tasks (PDFs, spreadsheets, etc.), activate the relevant skill first.
Multi-turn conversations: The agent remembers context, so you can build on previous responses.
Commands vs messages: Start with / for commands, anything else is sent to the AI agent.
Problem: command not found: agentao
Solution: Make sure the install completed: pip install agentao; or check that your Python scripts directory is on your PATH.
Problem: API key error
Solution: Make sure .env file exists and contains a valid OPENAI_API_KEY.
Problem: Tool execution fails Solution: Check file permissions and paths. Use absolute paths if relative paths don’t work.
/help command/skills commandsave_memory toolEnjoy using Agentao!