# Pylon MCP > Pylon connects your B2B support and CRM to any AI agent. Use this server within Slack or Teams to manage customer accounts, track open issues, and search knowledge bases without leaving your chat window. Your AI client handles the whole workflow—from listing all active tickets using list_issues to replying directly via reply_to_issue. ## Overview - **Category:** industry-titans - **Price:** Free - **Tags:** b2b-support, shared-channels, issue-tracking, account-management, post-sales-support ## Description **Pylon connects your B2B support and CRM directly to any AI agent you run.** You can manage customer accounts, track open issues, and search knowledge bases without ever leaving your chat window. Your AI client handles the whole workflow—it's like having a dedicated support specialist right here in Slack or Teams. *** **Handling Customer Issues:** If you need to see what's going on with support tickets, it starts with `list_issues`; that tool gives you a complete rundown of every open issue in the system. You can then dive deep into any single ticket using `get_issue` to pull the core metadata and current status. Need to know exactly why a customer opened the ticket? Run `get_issue_messages`, and your agent pulls the entire message thread history so you understand the whole context of their request. When you need to fix or change something, it's simple: you use `update_issue` to modify metadata or status on an existing ticket. Or, if a quick answer is all that's needed, `reply_to_issue` sends a direct reply, updating both the history and the issue's overall status. To start the process from scratch, your agent runs `create_issue`, which makes a brand-new support ticket using whatever data you provide. You can also check out all the tags used across issues by running `list_tags` to see every category name in use. **Managing Accounts:** You don't gotta guess who your clients are. Run `list_accounts`, and your agent generates a full list of every customer account Pylon manages. If you only need details on one client, `get_account` pulls all the specific profile information for that single account by its ID. **Searching Documentation:** Finding answers shouldn't mean opening ten tabs. Your agent can check which documentation sources are available using `list_knowledge_bases`, giving you a list of every connected knowledge base. To get actual articles, run `list_articles` to pull an overview and list of all the usable knowledge base documents so your AI client knows where to look for answers. **Putting It Together:** Your agent runs these tools together seamlessly. You can grab a client's profile with `get_account`, see if they have any open tickets using `list_issues`, check the whole conversation history via `get_issue_messages`, and then immediately reply to it using `reply_to_issue`—all without you ever switching applications. It keeps all the account data, ticket status, and knowledge base info right here in the chat thread. ## Tools ### create_issue Makes a brand new support issue ticket in Pylon using provided data. ### get_account Retrieves all specific details for one customer account by its ID. ### get_issue_messages Retrieves the entire message thread history for a specific support issue. ### get_issue Gets the main metadata and status of a single support issue ticket. ### list_accounts Generates a list of all customer accounts currently managed in Pylon. ### list_articles Provides an overview and list of available knowledge base articles. ### list_issues Generates a comprehensive list of all open support issues in the system. ### list_knowledge_bases Lists all connected knowledge bases available for searching documentation. ### list_tags Returns a list of every issue tag currently used in the system. ### reply_to_issue Sends a direct reply to an existing support issue ticket, updating its status and history. ### update_issue Modifies the metadata or status of a specific support issue ticket. ## Prompt Examples **Prompt:** ``` List all open issues in my Pylon account. ``` **Response:** ``` I've retrieved your issues. You have 3 open support tickets including 'Integration bug' and 'Billing question'. Would you like to see the messages for any of them? ``` **Prompt:** ``` Show me all open support issues assigned to the engineering team sorted by priority. ``` **Response:** ``` 18 open issues assigned to Engineering. Critical (3): "API rate limiting not working" (Acme Corp, 4 hours old), "Data sync failure" (TechFlow, 6 hours), "Authentication timeout" (Meridian, 2 days). High (7): includes payment processing delays, webhook delivery issues, and SDK compatibility bugs. Medium (5): feature requests and configuration questions. Low (3): documentation corrections. Average first response time: 23 minutes. SLA compliance: 94%. ``` **Prompt:** ``` Reply to the Acme Corp API rate limiting issue with a status update and estimated resolution time. ``` **Response:** ``` Reply sent to Acme Corp issue #ISS-4521. Message: "Hi team, we've identified the root cause of the rate limiting issue. A configuration change in our load balancer was incorrectly throttling requests below the documented limits. A fix has been deployed to staging and is being validated. ETA for production: within 2 hours." Status updated to: In Progress. Priority maintained: Critical. Internal note added for the engineering team with the specific config change details. ``` ## Capabilities ### Issue Lifecycle Management List all tickets using list_issues, retrieve full details for an issue with get_issue, create new issues via create_issue, and update or reply to existing ones using update_issue or reply_to_issue. ### Customer Account Retrieval List all customer accounts with list_accounts, or fetch specific client details for a single account using get_account. ### Communication History Analysis Fetch and analyze the full message history of any support issue via get_issue_messages to understand the context behind a customer's request. ### Knowledge Base Searching List available documentation sources using list_knowledge_bases, or retrieve specific articles using list_articles to answer questions instantly. ## Use Cases ### Handling an Escalation A client asks about a recent data sync failure. Your agent first runs list_issues to verify the ticket exists, then uses get_issue_messages to pull all past conversation details, and finally uses get_account to check their service tier before replying with precise information. ### Onboarding a New Client A new sales account needs support. You use list_accounts to confirm the client's profile exists, then run create_issue immediately to open the initial ticket, and finally use get_account to pull necessary billing details for the agent. ### Addressing a Billing Question The user asks about pricing. Your agent runs list_knowledge_bases and then uses list_articles to find the most current pricing document, summarizing it and using reply_to_issue to send the answer directly into the support ticket. ### Weekly Support Review The Ops manager needs a summary. They run list_issues and then filter by tag (list_tags) or status, getting a high-level report of all tickets requiring follow-up without ever touching the Pylon UI. ## Benefits - Instant Issue Visibility: Instead of logging into the ticketing dashboard, run list_issues or get_issue directly in your chat to see all open tickets assigned to a client. You know exactly where to start. - Full Context Retrieval: Running get_issue_messages pulls every message and comment on a ticket's history. This means you never have to ask 'What was the last thing they said?' again. - Zero-Click Account Lookups: Need to check if Acme Corp has changed their billing plan? Use get_account or list_accounts to pull that profile data right into your conversation, keeping the client in context. - Automated Responses: When you find an answer in the knowledge base (using list_articles), your agent can draft and send a complete reply using reply_to_issue, updating both the ticket and the customer's record. - Proactive Issue Creation: If a conversation reveals a problem that needs tracking—like a service outage—you don't waste time switching apps; you just use create_issue to open it right away. ## How It Works The bottom line is: Your AI agent does the work. You just talk to it in Slack or Teams. 1. Subscribe to the Pylon server and grab your API Token from your Pylon dashboard. 2. Plug that token into your preferred MCP-compatible client (like Claude or Cursor). 3. Tell your AI agent what you need—for example, 'What are all open issues for Acme Corp?' and let it run the necessary tools. ## Frequently Asked Questions **How do I check if a customer has existing issues using the Pylon MCP Server?** Run list_issues to get an overview of every open ticket. If you know the client, run list_accounts first, then use get_issue for that specific account's tickets. **Can I reply to a support issue using Pylon MCP Server?** Yes. Use the 'reply_to_issue' tool. Your agent sends the message and updates the ticket history in one step, so you don't have to manually log back into the system. **Is Pylon MCP Server only for Slack?** No. This server works with any AI client compatible with the Model Context Protocol (MCP), including Cursor, Claude, and others. It's designed to run in your preferred chat environment. **What if I need to check account data before listing issues?** You can run get_account first using the client ID, which validates the user context. Then you follow up with list_issues to filter only those tickets related to that specific account. **How do I use the `list_articles` tool with Pylon MCP Server to find documentation?** The server lists all available knowledge base articles for you. You simply ask your AI client a question, and it pulls relevant documentation directly from these indexed sources. **When should I use the `create_issue` tool with Pylon MCP Server?** Use this when an issue needs formal tracking in Pylon. You pass the required data—like customer name and description—and the agent instantly generates a new support ticket. **How do I change the status or metadata of a ticket using `update_issue` with Pylon MCP Server?** You execute the `update_issue` tool. Just provide the issue ID and the specific field you want to change, like setting the priority or adding an internal note. **Can I check all available tags using `list_tags` in Pylon MCP Server?** Yes, running `list_tags` retrieves every active issue tag in your system. This lets you ensure any new ticket is categorized with the correct metadata. **Can my AI automatically find the latest messages for a specific support issue?** Yes! Use the `get_issue_messages` tool with the Issue ID. Your agent will respond with the complete metadata for the conversation thread, including sender details and timestamps in seconds. **How do I find my Pylon API Token?** Log in to your Pylon dashboard, navigate to **Settings** > **API** (or `app.getpylon.com/settings/api`), and you will find or generate your unique secret token there.