# Typebot MCP > Typebot MCP Server lets your AI client manage complex conversational flows and chatbots. Need to check bot logic, deploy updates, or track leads? Connect this server to get full control of your Typebot workspace—all from a single chat interface. ## Overview - **Category:** productivity - **Price:** Free - **Tags:** chatbot-builder, conversational-forms, automation, visual-editor, lead-generation, open-source ## Description You gotta connect your AI client to this server and forget about clicking around in the Typebot dashboard. This thing lets you manage every single conversational flow and chatbot through natural language commands, right from your chat window. You get full control of your entire bot infrastructure without leaving your agent—it's pure power. Want to know where all your projects are? Start by calling `list_workspaces` so you can see a roster of every Typebot workspace you have access to manage. Once you've pinpointed the right space, you can drill down further using `list_folders`, which gives you a complete map of how your conversational projects are organized within that specific area. Need to know what bots live in there? You use `list_typebots` to pull up an inventory of every active chatbot. You can even filter this list by a workspace ID if you only want to check one section. If you need the full structural blueprint for any single bot, calling `get_typebot_details` retrieves all the metadata and the entire flow structure—you don't have to open up the editor just to review how it's built. Testing a bot before going live? You use `start_chat_session`. This command kicks off a simulated conversation, letting you test specific user inputs or complex conversational branches without risking your actual users. When you know the bot is ready for its audience, `publish_typebot` deploys the latest set of changes immediately, pushing it live instantly and skipping any manual steps. When you're done with a project—or if something went sideways—you can permanently wipe out an old setup using `delete_typebot`, ensuring that bot is gone for good. You also gotta track what the bots are actually doing; calling `list_typebot_results` gathers all the collected user responses and conversation outcomes, making it super simple to export data and analyze lead performance. Keep in mind that this server handles everything from organization to deployment and analysis. If you need a quick overview of your bot inventory or if you're checking how users interacted with a specific conversational flow, there are tools for those too. You can confirm the complete structure of any single Typebot using `get_typebot_details` before you deploy it, which helps you verify that all logic paths work as intended. This gives your agent command over every part of the Typebot lifecycle: listing what's available (`list_workspaces`, `list_folders`), checking on the bots themselves (`list_typebots`, `get_typebot_details`), simulating interactions for testing (`start_chat_session`), pushing updates live (`publish_typebot`), gathering performance data and leads (`list_typebot_results`), and even managing cleanup by removing old assets (`delete_typebot`). You've got complete oversight of your entire bot ecosystem, all from a single chat interface. ## Tools ### delete_typebot Permanently removes a specified Typebot from your account. ### get_typebot_details Retrieves the full structure and metadata for one specific Typebot, useful for reviewing its internal logic. ### list_typebots Lists all active conversational typebots, optionally filtering by workspace ID. ### list_folders Lists all organizational folders within a given workspace. ### list_typebot_results Retrieves and lists collected user responses for a bot, essential for lead analysis. ### list_workspaces Lists all accessible Typebot workspaces where you have access to manage bots and folders. ### publish_typebot Deploys the latest changes for a bot, making it live without manual steps. ### start_chat_session Starts a new, simulated conversation with a bot to test its flow or automated interaction. ## Prompt Examples **Prompt:** ``` List all typebots in my current workspace. ``` **Response:** ``` I've retrieved your typebots. You have 3 active bots: 'Lead Gen Flow', 'Customer Survey v2', and 'Support Assistant'. Which one would you like to publish or analyze? ``` **Prompt:** ``` Show me the collected results for the 'Lead Gen Flow' bot. ``` **Response:** ``` I've fetched the results. There are 15 recent submissions including leads from 'John Doe' (john@example.com) and 'Sarah Smith'. Shall I retrieve the detailed answers for any of them? ``` **Prompt:** ``` Publish the latest changes for bot 'bot_10293'. ``` **Response:** ``` Success! The latest changes for bot bot_10293 have been published to production. Your conversational form is now up-to-date at its public URL. ``` ## Capabilities ### List Bot Workspaces Fetch a list of all available Typebot workspaces so you know where your projects live. ### Get Bot Flow Details Retrieve the complete structure and metadata for a single bot, letting you review its logic flow without opening the editor. ### Start Test Chat Sessions Programmatically initiate a new chat session to test specific conversation branches or user inputs. ### Deploy Bot Updates Publish the latest changes for a specified bot ID, pushing it live immediately. ### Gather User Leads and Results List all collected user responses and conversation results, making lead export simple. ### Manage Bot Organization List or manage the folders and workspaces that keep your conversational projects structured. ## Use Cases ### QA Testing Bot Logic A QA tester needs to confirm that a multi-step survey bot correctly handles an unexpected user input (e.g., entering text when it expects a date). Instead of manually trying the edge case, they run `start_chat_session` through their agent. The agent executes the chat, and the output confirms if the bot logic breaks or handles the exception gracefully. ### Bulk Lead Export A marketing rep needs to gather all leads generated by a specific 'Q3 Campaign Bot' over the last month. The agent runs `list_typebot_results` for that bot, pulling 50+ submissions into a clean list format ready for immediate CRM import. ### Auditing Project Structure A PM joins the team and needs to know what bots exist across different departments. They run `list_workspaces` first, then use `list_typebots` filtered by a workspace ID to get a full inventory without needing admin access. ### Emergency Bot Update A critical bug is found in the main support bot. Instead of logging into Typebot and following a multi-step deployment wizard, the PM simply tells their agent to `publish_typebot` using the bot's unique ID. The fix goes live immediately. ## Benefits - Test flows instantly with `start_chat_session`. You don't have to manually interact with the bot in the UI; you can programmatically run chats to verify logic and user experience right through your agent. - Instantly deploy updates using `publish_typebot`. No more navigating to a 'Publish' button. Just ask your AI client, and the changes go live across all environments. - Stop hunting for leads. Use `list_typebot_results` to pull every user submission from any bot into one list, simplifying lead tracking and export processes. - Get a bird’s-eye view of everything with `list_workspaces`. Instead of clicking through dozens of folders, you ask the agent to list all projects across your organization's Typebot accounts. - Know exactly what a bot is doing with `get_typebot_details`. You can review the entire flow structure and metadata instantly without opening or navigating the visual editor. ## How It Works The bottom line is: you manage the entire Typebot lifecycle—from listing bots to deploying changes and analyzing leads—all via chat commands. 1. Subscribe to this server on Vinkius. 2. Provide your Typebot API Token (you find this in your account settings). 3. Ask your AI client to execute a command, like 'List all my workspaces' or 'Get details for bot X'. ## Frequently Asked Questions **How do I list all my bots using the Typebot MCP Server?** Run `list_typebots`. This command gives you a clean inventory of every bot across your connected workspaces. If you need to filter by location, you can also use `list_workspaces` first. **What if I need to test my Typebot flow before publishing?** Use the `start_chat_session` tool. This lets your agent programmatically start a chat with the bot, simulating real user interaction so you can debug complex flows without risking production data. **How do I get all lead submissions from my Typebot?** The `list_typebot_results` tool handles this. Just give the agent the bot ID, and it compiles a list of all collected user responses for you to analyze. **Can I use the Typebot MCP Server to see my organization's structure?** Yes. You can run `list_workspaces` to see every top-level workspace, and then use `list_folders` to drill down into the organizational groupings within that space. **How do I authenticate my connection when using `list_typebots`?** You must provide your Typebot API Token for authentication. Your AI client uses this token to authorize all calls, ensuring you only access data from your connected workspace. **What does the `get_typebot_details` tool actually retrieve?** This tool pulls the full structure and metadata for a specific bot. It's essential for reviewing the deep logic of a Typebot, allowing you to understand how different flow nodes connect. **What happens when I run the `publish_typebot` command?** The tool takes your latest saved changes and deploys them instantly to production. This means any updates made in your workspace are live and accessible at the bot's public URL. **Can I use `delete_typebot`? Is it safe?** The tool permanently removes a Typebot from your account. It handles all associated data, so make sure you have backed up or saved any necessary information before running the deletion command. **Can I publish a bot after making changes via AI?** Yes! Use the `publish_typebot` action and provide the unique bot ID. Your agent will instantly deploy the latest changes to the public URL. **How do I see the submissions or leads collected by a bot?** Run the `list_typebot_results` query with your Typebot ID. The agent will retrieve the complete history of user responses and collected data. **Is it possible to list bots from a specific workspace?** Absolutely. Use the `list_typebots` tool and provide the optional `workspace_id` to retrieve only the bots associated with that team area.