search-conversationsLABSCO SUMMARY
The support inbox as something you can ask about.
It does one thing: it finds conversations. That is usually the hard part — the filters cover date ranges, state, channel and read status, so 'everything unread from email since Monday' is a query rather than an afternoon.
What it is
An MCP server over the Intercom conversations API. One tool, search-conversations, with filters — and each result carries conversation details, contact information, response and reopen counts, and state and priority.
What you get
- Conversations filtered by creation date, with an operator and a UNIX timestamp
- Conversations filtered by last update the same way
- Filtering by source type, such as email or chat
- Filtering by conversation state, and by open and read booleans
- Contact information, response and reopen counts, state and priority attached to every result
Requirements
An Intercom API key, created in your Intercom workspace's developer settings and passed to the server as an environment variable. Clone, npm install, npm run build, then point your client at dist/index.js. Read access only — there is no way here to reply to a conversation or change its state, so it cannot touch your inbox.
Setup effort
Build from source — git clone https://github.com/fabian1710/mcp-intercom.git
USE CASES
Seeing what came in overnight
Filter by creation date with a timestamp and read back what landed.
Finding the ones nobody has answered
Filter by open and unread together, so the backlog is a list rather than a feeling.
Separating email from chat
Filter by source type to see how each channel is behaving.
Spotting threads that keep coming back
Pull conversations by state and read the reopen counts that come with them.
Reviewing a stretch of time for a report
Filter by update date between two timestamps and work from what returns.
1 tool
Listed in project docs. Your client may show a different list.
BEFORE YOU INSTALL
What to know first
Tools you get
1 tool named in the project's documentation.
Requires
INTERCOM_API_KEY in your environment
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
You clone and build this one yourself.
Server setup
git clone https://github.com/fabian1710/mcp-intercom.git
cd mcp-intercom
npm install
npm run buildMCP Clients
Claude Desktop
{
"mcpServers": {
"intercom": {
"command": "node",
"args": [
"<YOUR_PATH_TO_MCP_INTERCOM>/dist/index.js"
],
"env": {
"INTERCOM_API_KEY": "<YOUR_INTERCOM_API_KEY>"
}
}
}
}