# Intercom MCP > Intercom MCP connects your AI agent directly to your Intercom account, giving it full conversational control over your customer data. Your agent can list contacts, review full conversation histories, search across company records, and send messages—all without you ever leaving your chat interface. ## Overview - **Category:** talk-to-me - **Price:** Free - **Tags:** conversational-ai, helpdesk, customer-messaging, crm-integration, support-automation ## Description Manage all your customer communications from one place using this MCP connector. It gives your AI client direct access to the core functions of Intercom, allowing it to handle complex support inquiries or sales lead qualification through natural conversation. Instead of copying IDs or switching between tabs, you can ask your agent to look up a contact's full profile, check their history, and even reply directly. This dramatically speeds up everything from onboarding new users to handling urgent customer complaints. When you connect this MCP via Vinkius, your AI client acts like an internal expert, surfacing the exact information—like custom tags or company details—you need to keep the conversation flowing naturally. ## Tools ### create_contact Registers a brand new contact into Intercom’s system. ### create_conversation Starts a brand-new chat thread with a customer or prospect. ### get_contact_details Retrieves the full profile and custom data for a specific contact. ### get_conversation_details Pulls all historical messages from an existing conversation thread. ### list_admins Lists the team members who have administrative roles within the workspace. ### list_help_articles Retrieves a list of available help center articles for reference. ### list_contacts Displays an overview of all contacts registered in Intercom. ### list_conversations Shows a list of the most recent and currently open customer conversations. ### list_data_attributes Lists all custom data fields that can be attached to contacts or companies. ### list_segments Displays the different user segments defined in the workspace. ### reply_to_conversation Sends a response message to an ongoing conversation thread. ### search_contacts Searches for contacts using specific filters and criteria defined in Intercom. ## Prompt Examples **Prompt:** ``` Show all open conversations and the contact details for the most recent one. ``` **Response:** ``` 18 open conversations. Latest: Conv #4521 — 'Ana Costa' (ana@company.com) asking about pricing, 10 min ago, 3 messages. Contact: Ana Costa, Company: TechCorp (50 employees), Plan: Pro, Last seen: 2 min ago, Tags: enterprise-lead. Would you like to reply or search contacts? ``` **Prompt:** ``` Search for all contacts at acmecorp.com and list their companies. ``` **Response:** ``` 4 contacts at acmecorp.com: John Smith (CTO), Lisa Park (VP Eng), Mike Torres (Director), Sarah Chen (PM). Company: Acme Corp — 250 employees, Enterprise plan, $12K MRR, 4 active users, last activity today. Tags: enterprise, high-value. Would you like to send a message or view conversations? ``` **Prompt:** ``` Send a follow-up message to Ana Costa and list all team admins. ``` **Response:** ``` Message sent! ✅ To: Ana Costa. Content: 'Hi Ana, following up on your pricing question. Our Pro plan at $99/mo includes all features you mentioned. Happy to schedule a demo!' Team admins: 6. Sarah (Owner), Mike (Admin), Lisa (Admin), James (Member), Chen (Member), David (Member). ``` ## Capabilities ### Find and verify customer records Search for specific contacts using filters or retrieve a full profile including custom attributes. ### Review complete chat history Access and read the entire message thread of any conversation to understand the full context of a customer's issue. ### Initiate new conversations Start a completely fresh chat thread with a contact when needed. ### Send and reply to messages Write and send follow-up messages or replies directly within an active conversation. ## Use Cases ### Handling a complex billing question A customer asks about their usage plan. Your agent first runs `get_contact_details` to verify the customer's current plan and then uses `list_data_attributes` to pull up relevant custom data, allowing it to draft an accurate, personalized explanation. ### Qualifying a cold lead You need to know if a prospect works at a target company. Your agent runs `search_contacts` using the company domain and then uses `get_contact_details` on the results, providing immediate insights into their role (e.g., CTO or VP) for your next pitch. ### Following up after a demo After a product tour, you want to send follow-up materials. Your agent uses `list_contacts` to find the right person and then initiates a new chat using `create_conversation`, drafting a tailored message for them. ### Reviewing support trends You need an overview of recent issues. You prompt your agent to run `list_conversations` and ask it to summarize the top three topics, giving you immediate visibility into what customers are struggling with right now. ## Benefits - Instantly get customer context. Instead of manually checking a profile for tags or attributes, you can ask your agent to use `get_contact_details` and have it surface the exact information needed for your reply. - Never lose chat history again. You can prompt your agent to run `list_conversations` and then specify `get_conversation_details`, giving you immediate access to every message sent during a long troubleshooting session. - Improve outreach efficiency. Use `search_contacts` to filter leads by industry or custom tags, allowing your agent to draft hyper-specific emails before you even hit send. - Handle multi-step workflows faster. Your agent can combine tools—for instance, running `list_segments` first, then using that data to write a targeted message via `reply_to_conversation`. - Streamline internal operations. You don't need to switch tabs to check team roles; the agent can run `list_admins` and tell you exactly who has permission to do what. ## How It Works The bottom line is that it hands your AI client the keys to read, search, and write data inside Intercom. 1. First, subscribe to this MCP and provide your Intercom Access Token from the developer hub. 2. Next, point your preferred AI client (like Cursor or Claude) to this connector within Vinkius. This establishes secure access. 3. Finally, prompt your agent with a task—for example, 'Find all contacts who viewed pricing pages last week.' The agent executes the necessary tools and gives you an answer. ## Frequently Asked Questions **How does the Intercom MCP allow me to see a customer's full chat history?** You use the `get_conversation_details` tool. This function pulls every message, status update, and piece of information from that specific conversation thread so your agent can summarize it for you. **Can I search for a contact if I only know their company name using the Intercom MCP?** Yes. You use `search_contacts` and provide filters based on company details or custom attributes, allowing your agent to narrow down leads even without an email address. **What if I want to send a reply but don't know the conversation ID?** You first run `list_conversations` to see recent activity. Once you identify the correct thread, your agent can then use `reply_to_conversation` to send the message. **Does the Intercom MCP help with team management?** It does. You can run `list_admins` and `list_data_attributes`. These tools let your agent report on who has administrative power or what custom data fields are available in your account. **Is the Intercom MCP only for reading data?** No. It's read and write capable. You can use `reply_to_conversation` to send messages, and even `create_contact` if you need to log a brand new lead.