Submit

Multi-Agent Task Assistant with A2A & MCP-inspired Architecture

@ogulcanakca

Agent-to-Agent (A2A) communication protocol for inter-agent coordination and a Model Context Protocol (MCP)-inspired architecture for interacting with external tool servers.
Overview

What is the Article Assistant?

The Article Assistant is a multi-agent system designed to assist users in drafting articles and conducting web research using a custom Agent-to-Agent (A2A) communication protocol and a Model Context Protocol (MCP)-inspired architecture.

How to use the Article Assistant?

Users can interact with the system through a Streamlit web interface, where they can submit requests for article drafts or web research on specific topics. The system processes these requests through a Supervisor LLM for validation and then delegates tasks to specialized agents.

Key features of the Article Assistant?

  • Article Drafting: Automatically generates article drafts based on user-defined topics and styles, saving them to Google Cloud Storage.
  • Web Research: Conducts simulated web research and provides summarized findings.
  • Input Validation: Ensures user inputs are safe and correctly formatted before processing.
  • Asynchronous Processing: Handles tasks asynchronously, allowing for efficient management of multiple requests.
  • Microservice Architecture: Built using Docker for easy deployment and scalability.

Use cases of the Article Assistant?

  1. Drafting articles for blogs or academic purposes.
  2. Conducting research on various topics to gather information quickly.
  3. Automating content generation for marketing or educational materials.

FAQ from the Article Assistant?

  • What technologies are used in the Article Assistant?

The system is built using Python, FastAPI, Langchain, and Docker, leveraging Anthropic LLMs for various functionalities.

  • Is the Article Assistant free to use?

Yes, the Article Assistant is open-source and available for anyone to use.

  • How does the A2A communication work?

The A2A protocol allows different agents within the system to communicate and coordinate tasks effectively, ensuring efficient processing of user requests.

© 2025 MCP.so. All rights reserved.

Build with ShipAny.