Skip to content

Human Interaction

This example demonstrates how AI agents can interact with humans:

  • Using agent capabilities for human interaction
  • Setting up a human agent in the pool
  • Allowing AI to request human input when needed

How It Works

  1. We set up two agents:
  2. An AI assistant with can_ask_agents capability

  3. A human agent using the special "human" provider type

  4. When the AI assistant encounters a question it can't answer:

  5. It recognizes the need for human input
  6. Uses its can_ask_agents capability to interact with the human agent

  7. Incorporates the human's response into its answer

  8. The conversation might look like this:

Assistant: I need to check about Project DoomsDay's status. Let me ask the human.
Human: Project DoomsDay is currently in Phase 2, with 60% completion.
Assistant: Based on the human's input, Project DoomsDay is in Phase 2 and is 60% complete.

This demonstrates how to:

  • Enable AI-human collaboration
  • Control when AI can request human input
  • Integrate human knowledge into AI responses

Code

main.py

# /// script
# dependencies = ["agentpool"]
# ///


"""Example of AI-Human interaction using agent capabilities.

This example demonstrates:
- Using a human agent for interactive input
- AI agent querying human agent when unsure
- Using the can_ask_agents capability
"""

from __future__ import annotations

import os

from agentpool import AgentPool, AgentsManifest
from agentpool.docs.utils import get_config_path, is_pyodide, run


# set your OpenAI API key here
os.environ["OPENAI_API_KEY"] = os.environ.get("OPENAI_API_KEY", "your_api_key_here")


QUESTION = """
What is the current status of Project DoomsDay?
This is crucial information that only a human would know.
If you don't know, ask the agent named "human".
"""


async def run_example() -> None:
    # Load config from YAML
    config_path = get_config_path(None if is_pyodide() else __file__)
    manifest = AgentsManifest.from_file(config_path)

    async with AgentPool(manifest) as pool:
        # Get the assistant agent
        assistant = pool.get_agent("assistant")

        # Run interaction
        await assistant.run(QUESTION)

        # Print conversation history
        print(await assistant.conversation.format_history())


if __name__ == "__main__":
    run(run_example())

config.yml

# yaml-language-server: $schema=https://raw.githubusercontent.com/Million-mo/agentpool/refs/heads/main/schema/config-schema.json
agents:
  assistant:
    type: native
    model: openai:gpt-5-nano
    tools:
      - type: agent_cli
    system_prompt: |
      You are a helpful assistant. When you're not sure about something,
      don't hesitate to ask the human agent for guidance.

  human:
    model:
      type: input
    description: "A human who can provide answers"