Amazon Connect MCP

🚀 Amazon Connect MCP Server

An MCP (Model Context Protocol) server enabling AI assistants to interact with Amazon Connect contact centers, built with FastMCP and boto3.

🚀 Quick Start

Prerequisites

AWS CLI & Credentials

1. Install the AWS CLI.
2. Configure your credentials:

# If you don't have a profile yet, create one
aws configure --profile my-connect-profile

# You'll be prompted for:
# - AWS Access Key ID
# - AWS Secret Access Key  
# - Default region (e.g., us-east-1)
# - Output format (json)

3. Verify your setup:

# List available profiles
aws configure list-profiles

# Test connectivity
aws sts get-caller-identity --profile my-connect-profile

Your profile needs IAM permissions for Amazon Connect. See Amazon Connect required permissions.

Installation

# Clone the repository
git clone https://github.com/mundurragacl/amazon-connect-mcp.git
cd amazon-connect-mcp

# Create virtual environment
python3 -m venv .venv
source .venv/bin/activate

# Install
pip install -e .

Running the Server

# Local (stdio) - for Claude Desktop, Cursor, etc.
fastmcp run src/amazon_connect_mcp/server.py

# Or as a module
python -m amazon_connect_mcp.server

Cursor

Or manually add to .cursor/mcp.json:

{
  "mcpServers": {
    "amazon-connect": {
      "command": "python",
      "args": ["-m", "amazon_connect_mcp.server"],
      "env": {
        "AWS_PROFILE": "your-profile"
      }
    }
  }
}

Kiro CLI

Add to .kiro/settings/mcp.json (workspace) or ~/.kiro/settings/mcp.json (global):

{
  "mcpServers": {
    "amazon-connect": {
      "command": "python",
      "args": ["-m", "amazon_connect_mcp.server"],
      "env": {
        "AWS_PROFILE": "your-profile"
      }
    }
  }
}

For macOS with virtual environment, use the full path approach:

{
  "mcpServers": {
    "amazon-connect": {
      "command": "sh",
      "args": ["-c", "cd /path/to/amazon-connect-mcp && source .venv/bin/activate && python -m amazon_connect_mcp.server"],
      "env": {
        "AWS_PROFILE": "your-profile"
      }
    }
  }
}

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

Recommended (macOS/Linux):

{
  "mcpServers": {
    "amazon-connect": {
      "command": "/path/to/amazon-connect-mcp/.venv/bin/amazon-connect-mcp",
      "env": {
        "AWS_PROFILE": "your-profile"
      }
    }
  }
}

Alternative (if above doesn't work):

{
  "mcpServers": {
    "amazon-connect": {
      "command": "python",
      "args": ["-m", "amazon_connect_mcp.server"],
      "cwd": "/path/to/amazon-connect-mcp",
      "env": {
        "AWS_PROFILE": "your-profile"
      }
    }
  }
}

✨ Features

General Features

• 88 Tools covering all major Amazon Connect services.
• 91 Configuration Templates - Industry-specific templates for cases, views, routing, and more.
• Multi-Region Support - List instances across all AWS regions.
• Setup Wizard - Guided setup for new Connect instances.
• Infrastructure as Code - Generate CloudFormation templates.

MCP Capabilities

This server implements the full Model Context Protocol specification:

• Tools - 88 tools for interacting with Amazon Connect services (cases, contacts, analytics, AI, etc.).
• Prompts - Pre-configured prompts for common workflows like instance setup, case creation, and contact center configuration.
• Resources - Access to 91+ configuration templates as contextual resources that can be attached to conversations.
• Skills - Specialized capabilities including the Setup Wizard for guided instance configuration and IaC generation.

Tool Categories

Template Categories

💻 Usage Examples

List All Connect Instances

# Lists instances across all AWS regions
list_instances()

# Filter by region
list_instances(region="us-west-2")

Get Real-Time Metrics

get_current_metrics(instance_id="your-instance-id")

Create a Case

create_case(
    domain_id="your-domain-id",
    template_id="your-template-id",
    fields={"title": "Customer Issue", "priority": "High"}
)

Use Configuration Templates

# List available templates
template_list()

# Get a specific template
template_get(category="cases", name="general_support")

# Customize a template
template_customize(
    category="cases",
    name="billing_inquiry",
    overrides={"name": "Custom Billing Template"}
)

Setup Wizard

# Start guided setup
wizard_start_setup(
    use_case="ai_enhanced",  # basic, cases_enabled, ai_enhanced, full_enterprise
    instance_name="my-contact-center",
    region="us-east-1"
)

# Generate CloudFormation template
wizard_get_iac_template(
    use_case="basic",
    instance_name="my-cc",
    region="us-east-1"
)

📚 Documentation

• Amazon Connect API Features
• MCP Architecture Best Practices
• Templates & Views Research
• TODO: Templates Implementation

🔧 Technical Details

MCP Capabilities Reference

This server provides the full Model Context Protocol specification with tools, prompts, resources, and skills.

Tools (88 Available)

Core operations for Amazon Connect management:

1. list_instances - List all Connect instances across AWS regions.
2. describe_instance - Get detailed information about a specific instance.
3. get_current_metrics - Retrieve real-time contact center metrics.
4. create_case - Create a new customer support case.
5. search_cases - Search cases with filters and sorting.
6. cases_create_template - Create custom case templates.
7. contacts_start_outbound_voice - Initiate outbound voice calls.
8. config_create_queue - Create new contact queues.
9. profiles_search - Search customer profiles.
10. ai_query_assistant - Query Amazon Q in Connect for answers.

Prompts (Workflow Guidance)

Pre-configured prompts for common workflows embedded in templates:

1. Instance Setup - Step-by-step guidance for new Connect instance configuration.
2. Case Creation Workflow - Complete workflow from domain setup to case creation.
3. Customer Profile Integration - Associate Customer Profiles with Connect instance.
4. Queue Configuration - Create queues with hours of operation and routing.
5. Agent Workspace Setup - Configure agent views and screen pops.
6. Outbound Campaign Launch - Set up and start outbound calling campaigns.
7. Contact Flow Design - Build IVR flows with proper error handling.
8. Analytics Setup - Configure metrics collection and reporting.
9. AI Integration - Enable Amazon Q in Connect with knowledge bases.
10. Multi-Region Deployment - Deploy across multiple AWS regions.

Resources (91+ Templates)

Configuration templates accessible as contextual resources:

1. Cases Templates - 20 industry-specific case templates (healthcare, finance, retail, etc.).
2. Agent Views - 15 screen pop and disposition form templates.
3. Data Tables - 8 routing and configuration tables (holidays, SLAs, outages).
4. Hours of Operation - 6 schedule templates (24x7, business hours, healthcare).
5. Queue Configurations - Industry-specific queue setups.
6. Routing Profiles - Agent skill-based routing templates.
7. Contact Flows - 15 IVR flow templates by industry.
8. Evaluation Forms - 6 quality management templates.
9. Customer Profile Layouts - Unified customer view configurations.
10. CloudFormation Templates - Infrastructure as Code for instance deployment.

Skills (Specialized Capabilities)

Advanced capabilities for complex operations:

1. Setup Wizard - Guided instance setup with use case selection (basic, cases, AI, enterprise).
2. IaC Generator - Generate CloudFormation templates from configuration choices.
3. Template Customizer - Customize templates with overrides and industry-specific fields.
4. Layout Visualizer - Drag-and-drop interface for designing case layouts.
5. Multi-Region Discovery - Automatically discover instances across all AWS regions.
6. Domain Association - Properly sequence Customer Profiles and Cases integration.
7. Workflow Validator - Validate configuration dependencies before deployment.
8. Template Loader - Dynamic template loading with LLM guidance.
9. Session Management - Region-aware session handling for API calls.
10. Error Recovery - Automatic retry and fallback for common API errors.

Available Templates

Cases Templates (Industry-Specific)

Cases Layouts

• general_support_layout - Standard Z-formation layout.
• healthcare_layout - Patient-focused layout.
• financial_services_layout - Account and transaction layout.
• retail_ecommerce_layout - Order and shipping layout.
• insurance_claims_layout - Policy and claims layout.
• telecom_layout - Service and device layout.
• travel_hospitality_layout - Reservation and booking layout.
• technology_layout - Product and issue layout.
• utilities_layout - Service address and meter layout.

Agent Views (Industry-Specific)

Data Tables

Routing Templates

Hours of Operation:

• business_hours - M-F 8am-5pm.
• 24x7 - 24/7 operation.
• healthcare_hours - Healthcare with on-call support.
• financial_services_hours - Extended hours with 24/7 fraud line.
• ecommerce_hours - Extended hours with peak season config.
• utilities_24x7 - 24/7 with emergency prioritization.

Queue Configurations:

• general_support - Standard queue config.
• healthcare_queues - Healthcare department queues.
• financial_queues - Banking queues with fraud priority.
• ecommerce_queues - Retail queues with VIP support.
• telecom_queues - Telecom service queues.

Agent Routing Profiles:

• support_agent - General support agent profile.
• healthcare_profiles - Healthcare agent profiles.
• financial_profiles - Financial services profiles.
• ecommerce_profiles - E-commerce agent profiles.

Infrastructure as Code

• basic_instance.yaml - CloudFormation template for basic setup.

Architecture

┌─────────────────────────────────────────────────────────┐
│           amazon-connect-mcp (86 Tools)                 │
│                                                         │
│  TIER 1 - Core (Always Available):                     │
│  ├── list_instances (multi-region)                     │
│  ├── describe_instance, list_queues                    │
│  ├── get_current_metrics, search_contacts              │
│  └── create_case, get_case, search_cases              │
│                                                         │
│  TIER 2 - Domain Tools:                                │
│  ├── cases_*      (16 tools)                           │
│  ├── contacts_*   (8 tools)                            │
│  ├── config_*     (17 tools)                           │
│  ├── analytics_*  (5 tools)                            │
│  ├── profiles_*   (8 tools)                            │
│  ├── campaigns_*  (10 tools)                           │
│  └── ai_*         (8 tools)                            │
│                                                         │
│  WIZARD & TEMPLATES:                                   │
│  ├── template_list, template_get, template_customize   │
│  └── wizard_start_setup, wizard_get_iac_template       │
└─────────────────────────────────────────────────────────┘

Development

# Install dev dependencies
pip install -e ".[dev]"

# Run tests
pytest

# Type checking
mypy src/

Requirements

• Python 3.12+
• AWS credentials with Amazon Connect permissions
• FastMCP 2.x
• boto3

📄 License

MIT License - see LICENSE for details.

Contributing

Contributions welcome! Please read the contributing guidelines before submitting PRs.