This project provides an AI agent designed to function as an Executive Assistant (EA), automating email management, scheduling, and communication for busy professionals. It leverages LangGraph for orchestration and integrates with LLMs like OpenAI and Anthropic, targeting users who need to delegate email tasks.

How It Works

The EAIA agent operates as a stateful, multi-agent system orchestrated by LangGraph. It ingests emails, triages them based on configurable rules, drafts responses, and can interact with calendars. The core logic is modular, allowing customization of reflection, triage, calendar, tone, and drafting behaviors through specific Python files.

Quick Start & Requirements

• Install: pip install -e . within a Python 3.11 virtual environment.
• Credentials: Requires OpenAI API key, Anthropic API key, and Google API credentials (including Gmail API enablement and OAuth setup). LangSmith API key is also needed.
• Setup: Google OAuth setup can take 15-30 minutes.
• Docs: Hosted version documentation

Highlighted Details

• Orchestrated using LangGraph for complex agent workflows.
• Supports both local development and deployment on LangGraph Cloud.
• Highly configurable via config.yaml for email handling, scheduling preferences, and response styles.
• Modular design allows direct modification of reflection, triage, calendar, and drafting logic.

Maintenance & Community

The project is part of the langchain-ai organization, indicating active development and community support within the LangChain ecosystem.

Licensing & Compatibility

The README does not explicitly state a license. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

Google OAuth setup for non-Google Workspace accounts requires manual configuration as a test user, which can be a barrier. Production deployment requires a LangSmith Plus account for LangGraph Cloud.