Crisp MCP. Know the full context of every visitor interaction.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Crisp MCP handles real-time support communication and visitor context retrieval. Use this to list active chats, get full message transcripts for any session, or pull detailed profiles on website visitors.
It lets you automate responses and track customer history directly through your agent.
What your AI agents can do
Get conversation
Retrieves high-level summary information for a specific chat session.
Get messages
Lists the full, chronological content of messages within any given conversation thread.
Get visitor profile
Pulls detailed records about a specific person who visited your website.
See every current and past chat session running on your website.
Fetch the complete message history for a specific support thread.
Access personal and behavioral data about any known website user.
Programmatically send follow-up messages directly into an active chat session.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 Compatible Gemini Waiting for input…
Crisp MCP: 6 Tools for Customer Context
Use these tools to analyze the full spectrum of customer interactions, from listing every chat session to pulling granular details on specific visitor profiles.
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 Crisp on Vinkius019dd0d8get conversation
Retrieves high-level summary information for a specific chat session.
019dd0d8get messages
Lists the full, chronological content of messages within any given conversation thread.
019dd0d8get visitor profile
Pulls detailed records about a specific person who visited your website.
019dd0d8list conversations
Generates a list of all ongoing and archived support chats on the site.
019dd0d8list visitors
Provides an enumeration of every known person who has visited your website.
019dd0d8send message
Sends a direct, actionable message into an active chat conversation.
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 Crisp, 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 Crisp. 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 6 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Support agents spend hours clicking between chat views and CRM records.
Right now, if a customer asks about an issue, your support staff has to jump into the chat window to see what was said. Then they have to open another tab or system to check the visitor's profile—did they already buy something? Were they looking at pricing pages? They copy-paste context between three different screens just to answer 'What happened here?'
With this MCP, your agent pulls all that data automatically. It collects chat transcripts and visitor profiles in one go. You get a single, cohesive summary of the interaction history, so you can respond without ever leaving your workspace.
Get Visitor Profile: Know who they are instantly.
Before this, knowing if 'John' was a first-time visitor or a long-term client required manually searching through logs and checking status tags. It was slow, frustrating, and often inaccurate by the time an agent found it.
Now, calling get_visitor_profile gives your agent immediate access to their full history and current status. You're not just responding; you're engaging with a person whose entire journey is visible.
What you can do with this MCP connector
Stop bouncing between chat dashboards, email inboxes, and CRM records just to answer one question. This MCP connects all your multi-channel support data into a single stream of conversation context. You can instantly list every active web chat session or pull the full message history for any thread, giving you immediate insight into what the visitor talked about last week or five minutes ago.
Need to know who you're talking to? Fetch detailed profiles on visitors, including their activity and status. If your agent needs to update a billing system after support, it can chain this MCP with a different service—that's where Vinkius shines. Because we keep all credentials passing through a zero-trust proxy, you know your keys never sit on disk while building complex automations that span multiple platforms.
019dd0d9-096b-712e-bbda-01b35c74561d 
How Crisp MCP Works
- 1 First, you'll need your Plugin ID, Plugin Key, and Website ID from the Crisp dashboard. You pass those credentials to your AI agent.
- 2 Next, tell your agent what you need—for example, 'List all open support chats.' The MCP runs the request against the live Crisp data.
- 3 Finally, your agent gets back structured data, like a list of conversations or a full visitor profile, ready for you to use in your next prompt.
The bottom line is, instead of manually logging into Crisp, you just ask your AI client for the information it needs.
Who Is Crisp MCP For?
This MCP is for Ops Managers and Support Engineers who are tired of context switching between five different tabs just to answer a single customer question. It’s for anyone whose job requires synthesizing data from chat, email, and profile records instantly.
Uses the MCP to check visitor profiles and conversation transcripts immediately when a client calls in, ensuring they never start a call without full context.
Runs checks using list_conversations to quickly audit which support threads are stalled or unread across the platform.
Uses the MCP to pull visitor details and track website traffic summaries, allowing them to time personalized outreach messages precisely.
What Changes When You Connect
- Don't waste time hunting through threads. Use get_messages to instantly pull complete message histories for any session, giving your agent the precise background needed for a perfect response.
- Never start from scratch with a customer. The MCP allows you to fetch detailed visitor profiles using get_visitor_profile, so your agent knows their name, past activity, and status right away.
- Automate follow-up communication. If an issue is resolved, use send_message to keep the loop closed by sending immediate updates without human intervention.
- Keep track of everything that's happening on site. list_conversations gives you a real-time view of all open support chats, so nothing gets missed in high-volume periods.
- Gain full visibility into your customer interactions. Because this MCP runs on Vinkius, you get an audit trail for every tool call and data point that passes through the agent.
Real-World Use Cases
Handling a complex support ticket
A user asks their agent about a billing discrepancy. The agent first uses list_conversations to find the right thread, then get_visitor_profile to confirm account details, and finally gets_messages to review previous troubleshooting steps before drafting an accurate response.
Reacting to abandoned carts
A marketing analyst sees a high-value visitor who hasn't been active in days. They use list_visitors to find the record, and then send_message to trigger a personalized follow-up chat message offering help.
Auditing support volume
An ops manager needs to know how many chats were open at midnight last night. They use list_conversations to quickly pull the necessary status reports without manually checking the dashboard.
Onboarding a new team member
A trainer wants to demonstrate context gathering. They ask their agent to get_messages for a specific session, showing how easily the AI can summarize long threads for training purposes.
The Tradeoffs
Treating chat data as static text
Copy-pasting an entire week's worth of messages into a single prompt because the agent 'can't see the full context.' This creates massive token waste and confuses the AI.
→ Use get_messages to pull only the last 10 relevant exchanges. If you need all history, use get_visitor_profile first, as it aggregates key data points for better summarization.
Manually tracking visitor status
Maintaining a separate spreadsheet that has to be updated every time someone changes their VIP tag or visits the pricing page.
→ Call get_visitor_profile. This pulls real-time, accurate metadata directly from Crisp's system, keeping your data current without manual effort.
Sending messages with generic text
Just saying 'Hey' in a chat thread and hoping the customer responds. The message lacks context or next steps.
→ Use send_message after calling get_conversation. Reference specific details from the conversation history to make your outreach useful and actionable.
When It Fits, When It Doesn't
You should use this MCP if your core workflow involves synthesizing information across multiple, distinct communication channels—chats, emails, and user profiles. Specifically, you need to know who someone is (get_visitor_profile) and what they talked about recently (get_messages). Don't use it if you only need simple website analytics; those are better handled by dedicated traffic tools. If your goal is simply to write a generic FAQ article, don't use this MCP; instead, focus on using the agent to summarize transcripts into draft knowledge base articles.
Common Questions About Crisp MCP
How do I list all active chats using the get_conversation tool? +
You use list_conversations to see an overview of every chat. This gives you conversation IDs, which you then pass to get_conversation if you need more detail on a specific session.
Can I send messages using the send_message tool? +
Yes, send_message lets your agent reply directly to an active chat. It's perfect for automated follow-ups or confirming next steps without human intervention.
What data does get_visitor_profile provide? +
It pulls detailed records on the person, including their visit history and current status tags. This is better than just looking at a name because it provides context about their intent.
Is list_visitors reliable for finding old customer data? +
list_visitors gives you an enumeration of people who visited, which helps track volume and identity. For deep history, combine this with get_messages to pull the specific transcript needed.
If I need the full transcript for a session, how do I use the get_messages tool? +
The get_messages tool retrieves every message sent within a specific conversation thread. It gives you the complete history, allowing your agent to provide context far beyond just the latest activity summary.
What is the scope of conversations I can monitor using list_conversations? +
list_conversations provides an overview of all website support sessions. You can use this data to check statuses and get IDs for both active chats and recently closed threads.
When sending a message, what format does the send_message tool require? +
The send_message tool requires you to pass all data as a single JSON string. You must structure your payload precisely for it to correctly dispatch assistance to an active chat session.
Does list_visitors handle pagination when I need many people records? +
list_visitors returns batches of visitor data, and you will likely need to loop through the results. Your agent can process these pages sequentially to gather comprehensive lists of your user base.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.