Multi-Email Agent Toolkit

A collection of specialized email agents that act as intelligent tools - each with unique abilities. Forward emails to different agents and receive expert responses based on their domain expertise.

←Back to examplesCopy
email agents
multi-agent
email forwarding
545

The Multi-Email Agent Toolkit introduces a powerful pattern where users can interact with specialized AI agents by simply forwarding emails to them. Each agent is configured with a dedicated email integration and a unique set of abilities, effectively turning emails into a universal interface for accessing different AI capabilities.

This architecture demonstrates how email—the most ubiquitous communication tool - can become an intelligent routing layer for AI services. Users don't need to learn new interfaces or install apps; they simply forward their questions or requests to the appropriate agent's email address and receive expert responses directly in their inbox.

The blueprint includes three specialized agents, each with distinct expertise:

The Research Agent is equipped with web search and content fetching abilities, making it ideal for research tasks. Forward an email asking about market trends, technical concepts, or competitive analysis, and receive a well-researched summary in response.

The Content Agent specializes in writing and editing. Forward rough drafts, notes, or content briefs, and receive polished, professional content back. This agent can transform meeting notes into action items, expand bullet points into full paragraphs, or refine messaging for clarity.

The Data Agent is connected to external services through various abilities. Forward requests for weather information, stock data, or other real-time information, and receive accurate, up-to-date responses pulled from live APIs.

Each agent operates independently with its own email integration, allowing users to maintain separate conversation threads for different topics. The session duration can be configured to maintain context across multiple email exchanges within a conversation thread.

This pattern scales elegantly - add new agents for HR inquiries, IT support, project updates, or any domain that benefits from asynchronous, email-based interaction. Organizations can create internal "email APIs" where employees forward requests to specialized agents that handle routine queries, freeing up human experts for complex issues.

To extend this blueprint, consider adding abilities for calendar management, task creation, CRM updates, or document generation. You could also implement routing logic where a central agent analyzes incoming emails and forwards them to the appropriate specialist automatically.

Backstory

Common information about the bot's experience, skills and personality. For more information, see the Backstory documentation.

You are the Research Agent, an expert at finding, analyzing, and summarizing information from the web. When someone forwards you an email with a question or research request, you use your web search and content extraction abilities to gather comprehensive information. YOUR ROLE: 1. UNDERSTAND THE REQUEST - Carefully read the forwarded email to understand what information is needed - Identify key topics, questions, or areas to research - Note any specific sources or perspectives requested 2. CONDUCT RESEARCH - Use web search to find relevant, authoritative sources - Fetch and extract content from promising URLs - Cross-reference information from multiple sources 3. SYNTHESIZE AND RESPOND - Create a clear, well-organized summary of your findings - Include key facts, statistics, and insights - Cite your sources for credibility - Highlight any limitations or areas needing further research RESPONSE FORMAT: Always structure your email responses with: - A brief summary of the main findings - Detailed sections organized by topic - Source references at the end - Suggestions for follow-up questions if relevant Be thorough yet concise. Aim to provide actionable insights, not just raw information. The person forwarding you emails is busy— make their research tasks effortless. The current date is ${EARTH_DATE}.

Skillset

This example uses a dedicated Skillset. Skillsets are collections of abilities that can be used to create a bot with a specific set of functions and features it can perform.

  • 📄

    Web Search

    Search the web for information on any topic
  • ❎

    Fetch Web Page

    Extract and read content from a web page URL
  • 🤖

    Web Research

    Research topics to inform content creation
  • 📍

    Get Weather

    Fetch current weather and forecasts for any location
  • ❎

    Fetch Web Data

    Extract data from web pages and APIs

Secrets

This example uses Secrets to store sensitive information such as API keys, passwords, and other credentials.

  • 🔐

    OpenWeatherMap API Key

    API key for accessing OpenWeatherMap weather data

Terraform Code

This blueprint can be deployed using Terraform, enabling infrastructure-as-code management of your ChatBotKit resources. Use the code below to recreate this example in your own environment.

View Terraform Code

Copy this Terraform configuration to deploy the blueprint resources:

Next steps:

  1. Save the code above to a file named main.tf
  2. Set your API key: export CHATBOTKIT_API_KEY=your-api-key
  3. Run terraform init to initialize
  4. Run terraform plan to preview changes
  5. Run terraform apply to deploy

Learn more about the Terraform provider

A dedicated team of experts is available to help you create your perfect chatbot. Reach out via or chat for more information.

More Awesome Examples