Skip to main content
Every agent follows the same structure. Once you understand it, you can create any agent you need.

​
File Location

Agents are stored in ~/agents on your robot. The system automatically discovers Python files in this directory—no registration or configuration required. Add a file, and the agent appears. Remove it, and the agent disappears.

​
Core Interface

Every agent implements four methods:
MethodReturnsPurpose
idstrUnique identifier (snake_case).
display_namestrHuman-readable name shown in the app.
get_skills()List[str]Skills this agent can use.
get_prompt()strPersonality and behavioral instructions.
Optional methods:
MethodReturnsPurpose
display_iconstrPath to a 32x32 pixel icon.
get_inputs()List[str]Input devices to activate (for example: ["micro"]).
uses_gaze()boolEnable person-tracking eye movement.

​
Minimal Example

The simplest possible agent: 
from typing import List
from brain_client.agent_types import Agent

class MyAgent(Agent):
    @property
    def id(self) -> str:
        return "my_agent"

    @property
    def display_name(self) -> str:
        return "My Agent"

    def get_skills(self) -> List[str]:
        return []

    def get_prompt(self) -> str:
        return "You are a robot."
This agent loads and runs, but does very little—it has no skills and a minimal prompt.

​
Complete Example

A functional agent with skills, inputs, and a detailed prompt: 
from typing import List
from brain_client.agent_types import Agent

class HelloWorld(Agent):
    """Greets visitors with a friendly wave."""

    @property
    def id(self) -> str:
        return "hello_world"

    @property
    def display_name(self) -> str:
        return "Hello World"

    @property
    def display_icon(self) -> str:
        return "assets/hello_world.png"

    def get_skills(self) -> List[str]:
        return ["navigate_to_position", "wave"]

    def get_inputs(self) -> List[str]:
        return ["micro"]

    def get_prompt(self) -> str:
        return """
You are a friendly robot who greets people.

- Speak in a casual, warm tone
- If you don't see anyone, turn around to look for them
- When you see a person, wave and say hello
- Respond to what people say to you
"""

    def uses_gaze(self) -> bool:
        return True
This agent combines navigation and waving skills, microphone input, a friendly prompt, and gaze tracking for more natural interaction.

​
Writing Effective Prompts

The prompt determines how the robot behaves. Skills define what’s possible; the prompt defines what actually happens. A good prompt defines personality, goals, constraints, and strategy in plain language. Be specific. The AI interprets your prompt literally, and vague instructions produce inconsistent behavior.

​
Template

Copy and modify this template for new agents: 
from typing import List
from brain_client.agent_types import Agent

class MyCustomAgent(Agent):
    """One-line description of what this agent does."""

    @property
    def id(self) -> str:
        return "my_custom_agent"

    @property
    def display_name(self) -> str:
        return "My Custom Agent"

    @property
    def display_icon(self) -> str:
        return "assets/my_icon.png"  # Optional

    def get_skills(self) -> List[str]:
        return [
            "navigate_to_position",
            "wave",
            "turn_and_move",
        ]

    def get_inputs(self) -> List[str]:
        return ["micro"]

    def get_prompt(self) -> str:
        return """
Describe who the robot is and what it should do.
Be explicit about personality, goals, strategy, and constraints.
"""

    def uses_gaze(self) -> bool:
        return True
Save this as my_agent.py in ~/agents, and the system will load it automatically.