• LOGIC
  • USERS
  • HUMAN
  • CLEANUP

How to audit bot story logic with your AI

Connect ChatBot.com to your AI client to inspect conversational flows and user paths. You can pinpoint exactly where a bot deviates from its intended logic.

Ask AI about this page

Short answer

How can I check if my chatbot's logic is working?

Your AI pulls your existing story structures and recent user interactions to find gaps or errors. You'll get a clear view of where paths break or where users get stuck. This breakdown shows you how to find those issues.

Where every issue lands.

Audit outcomes

The AI categorizes what it finds in your ChatBot.com data.

LOGIC

Broken paths identified

The AI finds stories where the logic flow doesn't match the intended design. You get a list of specific story IDs that need fixing.

USERS

Interaction gaps found

The AI looks at how real people are actually talking to your bot. It flags interactions that hit dead ends or unexpected responses.

HUMAN

Manual review needed

Some edge cases require a human to step in and rewrite the dialogue. The AI flags these specific moments for your attention.

CLEANUP

Ready for deployment

Once you fix the flagged issues, the AI confirms the stories are structurally sound. You'll know which paths are safe to push live.

The workflow

What your AI does when the request arrives.

Your AI acts as a quality assurance tester by pulling data directly from your ChatBot.com account.

  1. Map the stories

    The AI first pulls a full list of all your existing conversational structures to understand the scope of your bot.

    list_chatbot_stories

  2. Inspect specific flows

    It dives into the individual steps and logic branches of a specific story to see how it's built.

    get_story_details

  3. Review user history

    The AI examines actual conversations to see if the logic holds up against real-world input.

    list_story_interactions

  4. Report discrepancies

    The AI compares the intended story details against the actual interaction logs to find where they diverge.

    get_story_detailslist_story_interactions

Try it

Copy these to start.

Use these prompts to begin your audit.

Starting points

Swap out the specific story names or interaction IDs for the ones you want to check.

ChatBot Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_2Z0qncNhPQ1ZiAdfbHS8L08HKBLoVsL20O1Ipg89/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — ChatBot capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "chatbotcom-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_2Z0qncNhPQ1ZiAdfbHS8L08HKBLoVsL20O1Ipg89/mcp"
    }
  }
}
Copy into chat04

  • List all my ChatBot.com stories so I can pick one to audit.

  • Check the details for the 'Welcome Flow' story and tell me if any branches are missing.

  • Look at the recent interactions for the 'Order Support' story and find where users got stuck.

  • Compare the logic in my 'Lead Gen' story with the actual user interactions to find errors.

  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Start here

Connect ChatBot.com once, then ask.

Just link your account once. Your credentials stay encrypted, and you can immediately start asking your AI to audit your work.

Connect ChatBot.com to your AI

FAQ

How this task behaves

  • 01

    Can the AI change my chatbot stories?

    No. The AI can only read your stories and interactions to report on them. You have to make any changes manually in ChatBot.com.

  • 02

    Can the AI delete old interactions?

    No. This MCP only has tools to list and get details. It cannot delete or modify any data in your account.

  • 03

    How does the AI know if a path is broken?

    It compares the structural logic defined in the story details against the actual conversation logs to see if the bot responded as expected.

  • 04

    Do I need to provide my login every time?

    No. You connect the ChatBot.com MCP once through Vinkius, and it stays ready for your AI client.

  • 05

    Can I audit a specific user's conversation?

    Yes, if you provide the specific story ID, the AI can pull the interaction history for that story.

  • More questions about ChatBot.com? See everything the ChatBot.com Connector can do

Connect ChatBot to Claude, Cursor, ChatGPT & more