Intercom MCP. Manage customer conversations and contacts from your agent.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Intercom MCP Server gives your AI agent direct access to your customer communication data. Manage support conversations, search contacts, and pull company details without leaving your agent interface.
You can list conversations, find specific contacts by email or name, and even reply to threads. It handles the full spectrum of customer support data, from initial lead enrichment to final resolution metrics.
What your AI agents can do
Assign conversation
Assign a conversation to a team member
Close conversation
Requires the admin ID who is closing it.
Close/resolve a support conversation
Create contact
Pass role (user/lead), email, and optional name. Returns the new contact ID.
Create a new contact/lead in Intercom
You search the customer database by email, name, company, or custom attribute, getting back a specific contact record.
You list or search conversation threads to get status, assignment, and the full history of customer interactions.
You pull full company information, including their current plan, usage metrics, and key attributes.
You pull metrics on response times and resolution counts to analyze team efficiency and support load.
You write a reply to a specific conversation thread, keeping the support history current.
You pull a list of every company record in the system, which can be used for bulk data checks.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 Compatible Gemini Waiting for input…
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Intercom on Vinkius019e9aa4assign conversation
Assign a conversation to a team member
019e9aa4close conversation
Requires the admin ID who is closing it. Close/resolve a support conversation
019e9aa4create contact
Pass role (user/lead), email, and optional name. Returns the new contact ID. Create a new contact/lead in Intercom
019d75baget contact
Retrieves the specific details for one customer contact.
019d75baget conversation
Fetches the full history and current status of a single conversation thread.
019d75balist admins
Lists all team members who use the Intercom platform.
019d75balist articles
Retrieves a list of available help center articles.
019d75balist companies
Pulls a list of every company record in Intercom.
019d75balist contacts
Retrieves a list of all contacts or leads in the system.
019d75balist conversations
Retrieves a list of all active and past conversations.
019e9aa4list segments
List all user segments
019d75balist tags
Gets a list of predefined tags used to categorize data.
019d75bareply to conversation
Sends a reply message into an existing conversation thread.
019d75basearch contacts
Searches the contact database using specific criteria like name or email.
019e9aa4update contact
Pass the contact ID and a JSON body of fields to update. Update an existing Intercom contact
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Intercom, then connect any of our 4,800+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,800+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Intercom. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 15 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Having to check 10 different tabs to piece together customer context is a time sink.
Right now, if you need to know what a customer is struggling with, you open Intercom. You check the main dashboard. You filter by date. You click into the conversation. Then you jump to the company profile to see their plan, and maybe you open a separate sheet to count how many support tickets they've filed. It’s a mess of tabs, clicks, and manual copy-pasting.
With the Intercom MCP Server, you just ask your agent. 'What's the status of Acme Corp?' Your agent runs the necessary tools—`search_contacts`, `get_company`—and spits out the full report in one answer. You get the context, immediately.
Use the Intercom MCP Server to manage conversations and contacts.
Before, if you needed to update a lead or respond to a query, you had to manually navigate to the conversation and type the reply. Then, if you needed to follow up, you had to copy the contact's name and email and paste it into a separate outreach tool.
Now, your agent handles the whole loop. It finds the contact using `search_contacts`, verifies their company status, and sends the reply using `reply_to_conversation`. The entire interaction stays within your agent's flow.
What you can do with this MCP connector
Your AI agent gets direct access to your customer communication data via the Intercom MCP Server. You can manage support conversations, search contacts, and pull company details without leaving your agent interface. You'll get the full picture of your customer base, from initial lead info to final resolution metrics.
Searching and Retrieving Contacts
You can use list_contacts to get a list of every contact or lead in the system. If you know who you're talking to, you can use search_contacts to narrow that down using specific criteria like name or email. To get all the juicy details on one person, you just hit up get_contact, which retrieves all the specific details for a single customer contact.
Handling Conversations
To see what's going on with customer support, you can use list_conversations to pull a list of all active and past threads. You can then drill down with list_conversations to get the full history and current status of a single conversation thread. You'll keep the support history current by sending a reply to an existing thread using reply_to_conversation.
Company and Team Data
When you need to know about the company behind the contact, you can hit up list_companies to pull a list of every company record. You can also get a full company profile, including their current plan, usage metrics, and key attributes. You'll see a list of all team members using the Intercom platform by running list_admins, and you can grab a list of predefined tags used to categorize data with list_tags.
019d75ba-c2cd-7058-94f0-50c189f30062 
How Intercom MCP Works
- 1 Subscribe to the Intercom integration on the Vinkius Marketplace.
- 2 Create a private app in the Intercom Developer Hub and grab the Access Token.
- 3 Ask your AI agent to search contacts, list conversations, or pull customer data. The agent uses the token to connect and execute the tool call.
The bottom line is, your AI client talks directly to Intercom using the provided credentials, letting you run CRM commands without opening a browser.
Who Is Intercom MCP For?
The Customer Success Manager who needs to know an account's health right now. The Support Agent who can't afford to switch tabs. The Sales Development Rep who needs instant context on a lead before making a call. This is for anyone whose job relies on context from customer history.
Looks up account health, usage patterns, and recent support interactions for key accounts, all via the AI agent.
Gets instant answers about customer history and conversation context without having to switch between multiple internal tools.
Pulls contact and company data to enrich leads and build better outreach strategies before a call.
What Changes When You Connect
- Instantly get contact context. Instead of opening a separate CRM tab, your agent uses
search_contactsto find a user by email or name and provides details immediately. - Stay in the chat. With
get_conversationandlist_conversations, your agent pulls the full support history, letting you respond without switching tools or screens. - Understand account health. Use
list_companiesto pull company profiles and usage metrics, letting you quickly see if a customer is on the right plan or if they're slipping into churn. - Improve team performance. Pull support analytics by calling tools like
list_conversationsto track resolution times and see which agents need coaching. - Enrich outreach data. Before you call a prospect, run
search_contactsto get the latest company details, making your pitch specific and relevant. - Close the loop. Use
reply_to_conversationto send a message directly from your agent, keeping the conversation thread active without manual copy/paste.
Real-World Use Cases
A support agent needs to know a customer's history.
A customer calls in. The support agent asks their agent, 'What's the history for sarah@startup.io?'. The agent runs search_contacts and get_conversation, pulling up Sarah's profile, her company's plan, and the last three support threads. The agent delivers this context instantly, and the agent drafts the reply using reply_to_conversation.
A sales rep needs to qualify a lead quickly.
A sales rep is prepping for a call. They ask their agent, 'Pull all company details for Acme Corp.' The agent runs list_companies and search_contacts, finding Acme's user count and plan. The rep uses this data to adjust their pitch on the spot.
A CSM needs to assess overall support load.
The CSM needs a quick snapshot of support status. They ask their agent, 'How many conversations are open right now?'. The agent runs list_conversations, providing a count, a breakdown of unassigned threads, and the oldest open conversation date.
An ops manager needs to audit team performance.
The manager asks their agent, 'Show me the resolution metrics for the last month.' The agent runs the appropriate tools to pull conversation counts and response times, allowing the manager to identify bottlenecks in the support process.
The Tradeoffs
Trying to manually check every customer record.
You open the Intercom dashboard, filter by date, click into 12 different threads, copy names, and paste them into a spreadsheet just to track who talked to whom. It takes hours and you miss context.
→
Ask your agent to run list_contacts and list_conversations to pull the raw data. Then, use search_contacts with specific criteria to narrow the focus down to the records you care about.
Forgetting to check company plan details.
You assume a lead is on the Enterprise plan because they talk to sales. You don't check the actual plan, and your outreach sounds wrong, scaring them off.
→
Always run get_contact or search_contacts first. This gives you the company profile data, letting you verify the actual plan tier before you commit to a pitch.
Treating the AI as a simple search engine.
You ask the agent, 'What was Sarah's last message?' and it just lists her messages. You can't tell if the conversation is closed or who owns it.
→
Ask the agent to use get_conversation or list_conversations. This gives you the full thread and the current status, which is the critical context you need.
When It Fits, When It Doesn't
Use this if you need to manage customer context or operational data from Intercom. Specifically, if your workflow requires knowing who a user is (search_contacts), what their company status is (list_companies), or what the current state of a chat thread is (get_conversation). Don't use this if you just need general web knowledge or need to write a standalone article; for that, use a general knowledge model. If you need to run a bulk, irreversible action (like deleting records), look for dedicated API tools, because this server is focused on reading and communicating.
If your goal is to build a multi-step workflow (e.g., find contact -> get company details -> reply), this server is built for that. If you only need to call one tool in isolation, you might be better off with a specialized, single-function API wrapper, but this suite gives you the full context.
Common Questions About Intercom MCP
How do I use the `search_contacts` tool with Intercom MCP Server? +
You tell your agent to search contacts by providing criteria like an email address or a name. The agent executes search_contacts and returns the contact record, including company and plan details.
Can I use `list_companies` to see all my customer plans? +
Yes. Running list_companies pulls a list of every company record in Intercom, allowing you to review which plan tiers are in use across your entire customer base.
What's the difference between `list_contacts` and `search_contacts`? +
Use list_contacts if you need a raw dump of every contact. Use search_contacts when you know specific criteria (like a name or email) and need the data for a specific person.
How do I make the agent respond to a thread using `reply_to_conversation`? +
You instruct the agent to reply, providing the target conversation ID and the text you want to send. The agent executes the tool and sends the message on your behalf.
Do I need to use `get_contact` for every user lookup? +
No. get_contact gets a specific record when you know the ID. If you're starting from scratch, search_contacts is usually better because it lets you search by name or email first.
What parameters does `get_conversation` need to pull a specific chat thread? +
You need to provide the conversation ID. Once you have the ID, the tool returns the full transcript, including all messages, participants, and status updates for that specific thread.
Can `list_conversations` filter by conversation status or date range? +
It allows filtering by status, so you can specify if you only want open, closed, or snoozed conversations. You can also filter by the last updated date to check for recent activity.
How do I use `search_contacts` to find leads associated with a specific company? +
You pass the company name or ID as a criteria. The tool then filters the contact list, giving you all associated contacts and their specific details within that company structure.
How do I get my Intercom Access Token? +
Go to the Intercom Developer Hub at app.intercom.com/a/developer-hub/. Click New app, give it a name, and select the workspace to install it in. Once created, click on your app, then navigate to Authentication in the left sidebar. Your Access Token will be displayed on this page. Configure the required permissions (scopes) — such as Read/Write People, Conversations, or Companies — and click Save. Copy the token and paste it below.
Can I look up a customer's full conversation history instantly? +
Yes. Just ask your AI agent 'Show me all conversations with john@acme.com' and it retrieves every conversation — open, closed, and snoozed — including message content, agent assignments, tags, and resolution times. Perfect for getting full context before a customer call.
Can I track support team performance metrics? +
Yes. Your AI agent can pull conversation counts, average response times, resolution rates, and open ticket volumes — giving managers instant visibility into team workload and SLA compliance without building custom reports or opening the Intercom analytics dashboard.
Is my customer data secure? +
Your Access Token is a private app token that only accesses data in your own Intercom workspace. It respects your workspace's permission model — the token only has access to the scopes you explicitly grant. All API calls are made over HTTPS, and the token is never shared with third parties.
Multi-server workflows that include Intercom MCP
MCP Servers for Automated Customer Onboarding
One AI agent handles CRM, scheduling and messaging , no triggers, no webhooks
MCP Servers That Predict Subscriber Churn
A customer downgraded from Enterprise to Starter yesterday and nobody noticed , their 6 support conversations in the last month should have been the warning
MCP Workflow for Smart Support Ticket Escalation
Your AI agent triages support tickets, deduplicates escalations and logs everything , hands off
Predict Agency Client Churn Using MCP Servers
Your biggest client has not opened the last 3 emails, their Intercom conversations dropped 80% in 2 months, and the account manager finds out they are leaving when the cancellation email arrives , not 60 days earlier when the signals were already screaming
Score Customer Health Using MCP Servers
Support says the customer is happy, Sales says the deal is solid, but nobody checked that they opened 14 tickets last month and usage dropped 60%
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.