An MCP-compliant server that enables AI assistants to access and analyze Intercom customer support data. It provides robust search capabilities across conversations and tickets, with advanced filtering options such as by customer, status, date range, and keywords. It can search email content even when there is no associated contact, and relies on Intercom's server-side search API for efficient data retrieval. The server exposes a set of MCP tools (endpoints) including list_conversations, search_conversations_by_customer, search_tickets_by_status, and search_tickets_by_customer, allowing developers to build AI-assisted workflows that retrieve data within a configurable date window (the system notes that date ranges are typically limited to seven days). The project ships with a Glama-optimized Docker image and a lighter standard variant, and can also be installed via npm. Configuration for Claude Desktop is supported, with a sample claude_desktop_config.json showing how to supply INTERCOM_ACCESS_TOKEN. Implementation details and parameter mappings are documented in src/services/INTERCOM_API_NOTES.md. The license is Apache 2.0, and the project remains an independent integration, not officially affiliated with Intercom. It targets AI assistants and developers enabling MCP-enabled interfaces to query Intercom data.