MCP server for Intercom integration
Project description
Intercom MCP Server
A Model Context Protocol (MCP) server for Intercom integration. This server provides tools for interacting with Intercom, including managing conversations, contacts, companies, and help center articles.
Features
- Conversation Management: List, view, and reply to customer conversations
- Contact Management: Create, search, and view user contacts
- Company Management: List and view companies
- Help Center: Access articles and content
- Team Management: View teams and admins
- Resources: Direct access to Intercom data
- Prompts: Templates for common Intercom workflows
Installation
pip install mcp-intercom
Configuration
Set the following environment variables:
export INTERCOM_BASE_URL="https://api.intercom.io" # Default value, usually doesn't need changing
export INTERCOM_ACCESS_TOKEN="your_access_token" # Preferred authentication method
# OR
export INTERCOM_API_KEY="your_api_key" # Alternative authentication method
You'll need to create an Intercom access token with appropriate permissions from your Intercom Developer Hub.
Usage
Starting the server directly
mcp-intercom
Using with Claude Desktop
Add the following to your claude_desktop_config.json file:
"mcp-intercom": {
"command": "uvx",
"args": [
"mcp-intercom"
],
"env": {
"INTERCOM_ACCESS_TOKEN": "your_access_token"
}
}
Replace the environment variables with your actual Intercom credentials.
Available Tools
- list_conversations: List recent conversations in Intercom
- get_conversation: Get details of a specific conversation
- reply_to_conversation: Reply to a conversation as a comment or note
- list_contacts: List contacts in Intercom
- get_contact: Get details of a specific contact
- create_contact: Create a new contact
- list_companies: List companies in Intercom
- get_company: Get details of a specific company
- list_articles: List help center articles
- get_article: Get details of a specific article
- search_contacts: Search for contacts with a query
Available Resources
- intercom://conversations: List of recent conversations
- intercom://contacts: List of recent contacts
- intercom://teams: List of teams in your Intercom workspace
- intercom://admins: List of admin users in your Intercom workspace
- intercom://companies: List of recent companies
Available Prompts
- create_contact: Template for creating a new contact
- reply_to_conversation: Template for replying to a conversation
- search_knowledge_base: Template for searching Intercom's knowledge base
Example Usage
With Claude or another AI assistant, you can ask natural language questions like:
- "What are my recent Intercom conversations?"
- "Show me details for contact with email john@example.com"
- "Reply to conversation abc123 saying 'Thanks for reaching out!'"
- "Create a new contact with email jane@example.com and name Jane Smith"
Development
To contribute to this project:
git clone https://github.com/yourusername/mcp-intercom.git
cd mcp-intercom
pip install -e .
Version
0.0.1
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file mcp_intercom-0.0.2.tar.gz.
File metadata
- Download URL: mcp_intercom-0.0.2.tar.gz
- Upload date:
- Size: 5.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.6.13
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
ce71362f94d3178f5b3eecc23514f8c583bc565f788d2075dabd75fcf73a035d
|
|
| MD5 |
0e1c82cd1195c6ab889b1bf5690f30d5
|
|
| BLAKE2b-256 |
7b33ceb14a48f40eaff67127916e917c2fb529f67d2071598ea45ab6b67935e0
|
File details
Details for the file mcp_intercom-0.0.2-py3-none-any.whl.
File metadata
- Download URL: mcp_intercom-0.0.2-py3-none-any.whl
- Upload date:
- Size: 6.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: uv/0.6.13
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
4771aa2446169d64a8499c63ebc99d3f38bdb0c85a559656545af9981821c612
|
|
| MD5 |
8b455b2c8500209e47b3708db1674b2d
|
|
| BLAKE2b-256 |
2a0c8d541321ffc45c3bd55d829f4d8d1a4e9c9a8bd663d42259f4fe23520f9b
|