Once you’ve discovered agents and configured your environment, running them is simple.

You can launch and interact with agents using the graphical interface or the command line. To open the GUI, run:

beeai ui

Not all agents support GUI interaction.

​
Quickstart Examples

​
Input Options

Different agents may have different input expectations. Always refer to the agent’s documentation via beeai info <agent-name> for guidance.

​
Interactive Mode

Opens a session where you can send multiple messages.

beeai run chat

​
Direct Text Input

Run an agent with a single message or prompt.

beeai run gpt-researcher "Research the latest developments in quantum computing"

​
Multi-line Input

Great for sharing code snippets.

beeai run aider "
Review this Python function and suggest improvements:

def calculate_fibonacci(n):
    if n <= 1:
        return n
    return calculate_fibonacci(n-1) + calculate_fibonacci(n-2)
"

​
File Input

Send the contents of a file as input to an agent.

beeai run chat < my_file.txt

​
Pipe Between Agents

Send output from one agent into another.

beeai run gpt-researcher "Latest developments in renewable energy storage" | beeai run chat "Summarize this research in 3 key bullet points"

​
Output Handling

​
Save to File

Save the agent’s output to a file:

beeai run chat "Write a technical blog post about how large language models work" > ~/post.md

​
Append to File

Append output to an existing file:

beeai run chat "Add a section on the challenges and limitations of large language models" >> ~/blog.md

Make sure the path points to an existing file.