# Atlas MCP MCP > Atlas MCP lets your agent manage everything in Atlas.so, connecting directly to your support data without opening a browser. You can list all open tickets, get deep details on specific customers or articles, and even create new tickets right from your chat context. It handles customer history, team rosters, and knowledge base lookups—all through natural language commands. ## Overview - **Category:** productivity - **Price:** Free - **Tags:** ticketing-system, customer-service, knowledge-base, conversation-metadata, support-automation ## Description It's basically giving your AI agent full access to the Atlas support platform's backend data. Instead of having to copy IDs or navigate complex dashboards, you just ask it what you need. You can pull up a specific customer’s file, review every ticket they’ve ever opened, and check which help articles apply to their problem—all in one chat session. This makes your agent incredibly useful for anything from onboarding new hires to auditing support processes. If you're looking for the core services of major platforms like Zendesk or Freshdesk, this MCP gives you that access, allowing your agent to act as if it’s logged into the system itself. You connect everything through Vinkius, making Atlas data available instantly alongside other tools in your catalog. ## Tools ### create_ticket Automatically generates a brand-new support ticket record within Atlas.so. ### get_account_check Verifies the connection status between your AI agent and your Atlas account credentials. ### get_customer Retrieves a full profile summary for one specific customer by name or ID. ### get_ticket Fetches all conversation history and metadata associated with a single ticket number. ### list_articles Searches and returns a list of help center articles based on keywords you provide. ### list_customers Lists all customers in your Atlas account, allowing filtering by criteria like email or ID. ### list_tickets Retrieves a summary list of all active support tickets currently open in the system. ### list_users Shows you a roster of all current team agents and users who handle support requests. ## Prompt Examples **Prompt:** ``` List all active support tickets in Atlas. ``` **Response:** ``` I've retrieved your active tickets. You have 10 open issues, including 'Login Problem' and 'Billing Inquiry'. ``` **Prompt:** ``` Show me the details for ticket ID 'tick_12345'. ``` **Response:** ``` Ticket 'tick_12345' is titled 'Payment Failed'. The current status is 'Open' and it was created by 'John Doe' (john@example.com). ``` **Prompt:** ``` Find all help articles related to 'Pricing'. ``` **Response:** ``` I've found 3 articles related to pricing: 'Subscription Plans', 'Annual vs Monthly Billing', and 'Refund Policy'. ``` ## Capabilities ### Audit and Create Tickets Lists all active support tickets or generates a brand new ticket directly from conversation context. ### Manage Customer Records Retrieves specific customer details, including their internal ID, name, and email address. ### Access Knowledge Base Articles Searches and lists articles from your help center based on keywords or topics. ### Review Team Capacity Generates a list of current team users, letting you understand who is available to handle support requests. ### Get Specific Ticket Details Pulls up the full metadata and conversation history for any given ticket ID. ## Use Cases ### Customer needs immediate help. A user asks, 'Why is my payment failing?' Your agent runs `list_articles` for 'Billing' and cross-references the customer using `get_customer`, then uses that info to create a new ticket via `create_ticket` if no article helps. ### Manager needs an overview of today’s workload. A manager prompts, 'What are our top 5 open issues?' The agent calls `list_tickets`, providing a summary list and the IDs for the most critical items needing review. ### Agent needs context on an old issue. An agent receives a chat about 'Error Code 403'. They immediately call `get_ticket` with the ID provided in the chat, retrieving all historical notes and previous conversations to keep the customer updated. ### Identifying team bottlenecks. A support lead asks, 'Who is available right now?' The agent calls `list_users`, showing who is currently logged on and free to take new cases, ensuring efficient resource allocation. ## Benefits - Stop switching tabs. Instead of manually checking ticket statuses or pulling up customer profiles in separate browser windows, your agent can handle all that data retrieval—using `list_tickets` and `get_customer`—and present it back to you immediately. - Audit team capacity on demand. If you need to know who’s free before assigning a complex issue, simply call `list_users`. It gives you an instant overview of your support staff without navigating the internal dashboard. - Cut down response time with knowledge base access. When a customer asks about billing, instead of guessing, prompt the agent to run `list_articles` to pull up the exact official documentation page for that topic. - Automate initial issue logging. If you have a conversation where multiple problems are mentioned, don't waste time writing it all out; just use `create_ticket` and let the agent file the complaint for you. - Deep dive into history. Need to know why a customer complained last month? Use `get_ticket` with the specific ID. You get every piece of conversation metadata without digging through old emails. ## How It Works The bottom line is: you talk naturally, the agent does the API work, and you get the answer immediately. 1. Your agent first uses `get_account_check` to verify that your Atlas.so API credentials are correctly connected. 2. Next, you prompt the agent with a specific request—for instance, 'Show me all open tickets for this billing issue'—and it translates that into structured data calls. 3. The MCP processes these tools and returns clean metadata (e.g., ticket status, customer name) directly to your chat window. ## Frequently Asked Questions **How does the Atlas MCP use the `list_tickets` tool?** The agent uses `list_tickets` to pull a summary of all support issues. This lets your client quickly see how many tickets are open and what their general topics are, without needing to look at every single record. **Can I use the Atlas MCP to find customer info? (get_customer)** Yes, you can. Running `get_customer` lets your agent pull up all key details for a specific person—like their name and account ID—which is crucial context before tackling an issue. **What if I want to file a new ticket? (create_ticket)** You just prompt the agent with the necessary details. It uses `create_ticket` behind the scenes, filing the support request for you directly in Atlas without requiring manual form submissions. **Does the MCP help me find relevant articles? (list_articles)** Absolutely. The agent runs `list_articles` and searches your entire knowledge base using plain language, returning a list of documentation that directly addresses the user's query. **How do I verify that my Atlas connection is set up correctly using `get_account_check`?** The `get_account_check` tool immediately verifies your API token and Atlas credentials. It confirms connectivity status, letting you know right away if there are any setup issues before running deeper reports. **If I need a complete list of all my registered users, what does the `list_users` tool do?** The `list_users` tool pulls a directory of every agent on your team. This helps you monitor overall support capacity and understand who is available to handle incoming requests. **I need a full list of all my clients, not just one. How do I use `list_customers`?** `list_customers` retrieves the entire database roster of your clientele. This functionality lets you pull comprehensive reports or process data for mass communications. **I only have a ticket ID and need all its metadata. How can I use `get_ticket`?** The `get_ticket` tool pulls the complete, detailed record for one specific ticket ID. This is ideal when you need to deep-dive into a single conversation's history or status. **How do I find my Atlas.so API Token?** Log in to your Atlas dashboard, go to **App Configuration > Data > API**, and you will find your API Token there. **Can I see help center articles via this server?** Yes, use the `list_articles` tool to retrieve a list of articles from your Atlas help center. **How are customer IDs handled?** Atlas uses unique internal IDs for customers. You can discover these IDs by using the `list_customers` tool or searching for a specific customer by email.