Skip to content

Latest commit

 

History

History
79 lines (56 loc) · 2.82 KB

README.md

File metadata and controls

79 lines (56 loc) · 2.82 KB

Marvin Contact Extractor Agent (A2A Sample)

This sample demonstrates an agent using the Marvin framework to extract structured contact information from text, integrated with the Agent2Agent (A2A) protocol.

Overview

The agent receives text, attempts to extract contact details (name, email, phone, etc.) into a structured format using Marvin. It manages conversational state across multiple turns to gather required information (name, email) before confirming the extracted data. The agent's response includes both a textual summary/question and the structured data via A2A.

Key Components

  • Marvin ExtractorAgent (agent.py): Core logic using marvin for extraction and managing multi-turn state via a dictionary.
  • A2A AgentTaskManager (task_manager.py): Integrates the agent with the A2A protocol, managing task state (including streaming via SSE) and response formatting.
  • A2A Server (__main__.py): Hosts the agent and task manager.

Prerequisites

  • Python 3.12+
  • uv
  • OPENAI_API_KEY (or other LLM provider creds supported by pydantic-ai)

Setup & Running

  1. Navigate to the Python samples directory:

    cd samples/python/agents/marvin

  2. Set an LLM provider API key:

    export OPENAI_API_KEY=your_api_key_here

  3. Set up the Python environment:

    uv venv
source .venv/bin/activate
uv sync

  4. Run the Marvin agent server:

    # Default host/port (localhost:10030)
MARVIN_DATABASE_URL=sqlite+aiosqlite:///test.db MARVIN_LOG_LEVEL=DEBUG uv run .

# Custom host/port
# uv run . --host 0.0.0.0 --port 8080

    Without MARVIN_DATABASE_URL set, conversation history will not be persisted by session id.

  5. In a separate terminal, run an A2A client (e.g., the sample CLI):

    # Ensure the environment is active (source .venv/bin/activate)
cd samples/python/hosts/cli
uv run . --agent http://localhost:10030 # Use the correct agent URL/port

Extracted Data Structure

The structured data returned in the DataPart is defined as:

class ContactInfo(BaseModel):
    name: str = Field(description="Person's first and last name")
    email: EmailStr
    phone: str = Field(description="standardized phone number")
    organization: str | None = Field(None, description="org if mentioned")
    role: str | None = Field(None, description="title or role if mentioned")

with a validator to render things nicely if you want and maybe serialize weird things.

Learn More