{"id":16244,"date":"2026-07-24T14:03:21","date_gmt":"2026-07-24T19:03:21","guid":{"rendered":"https:\/\/flevy.com\/blog\/?p=16244"},"modified":"2026-07-24T14:06:36","modified_gmt":"2026-07-24T19:06:36","slug":"how-to-extend-your-ai-chatgpt-claude-copilot-with-the-full-flevy-knowledge-base","status":"publish","type":"post","link":"https:\/\/flevy.com\/blog\/how-to-extend-your-ai-chatgpt-claude-copilot-with-the-full-flevy-knowledge-base\/","title":{"rendered":"How to Extend Your AI (ChatGPT, Claude, Copilot) with the Full Flevy Knowledge Base"},"content":{"rendered":"

\"\"Most business professionals now use AI for strategic analysis and for structuring business problems. The answers are pretty good, but not great. They still do not compare to the level of analysis you would get from a real 100+ slide consulting deliverable.<\/p>\n

There is a specific reason for that gap. AI models are trained on the public internet and the consulting knowledge with genuine depth was never published there. The proprietary frameworks and methodology decks, client-ready deliverables, and the accumulated intellectual property from thousands of engagements were built in-house at the global strategy firms and deliberately kept there. That material never entered any training data, so your AI has never seen it.<\/p>\n

The gap is not intelligence. It is access.<\/p>\n

Flevy is where that caliber of material has been published. Since 2012, former McKinsey consultants and Fortune 100 executives have contributed over 10,000 frameworks, templates, deliverables, toolkits, and models, alongside 8,000 management case studies. These documents are the same caliber as the materials traditionally kept in-house at the global strategy firms.<\/p>\n

You can now connect that entire library directly to your AI assistant through the Flevy MCP<\/a>. This guide walks through how.<\/p>\n

What Is the Flevy MCP?<\/h2>\n

MCP stands for Model Context Protocol.<\/a> It is an open standard that lets an AI assistant connect to a live external data source and pull real information mid-conversation, instead of relying only on what it memorized during training.<\/p>\n

The Flevy MCP is our implementation of that standard. Once connected, your AI assistant can search Flevy’s full library on its own, retrieve document details including the author and their credentials, look inside a presentation slide by slide, and surface relevant case studies. Every result links back to the original document on Flevy.<\/p>\n

The connector is read-only, and the endpoint is always the same: https:\/\/flevy.com\/mcp<\/a><\/strong><\/p>\n

How to Connect It<\/h2>\n

The Flevy connector works with any MCP-compatible AI client.<\/p>\n

A quick note before the steps. These AI tools update their interfaces frequently, so the exact menu names below may have shifted by the time you read this. If something does not match what you see, check the provider’s own connector documentation. The endpoint URL never changes.<\/p>\n

Claude Desktop<\/h3>\n
    \n
  1. Click your name, then Settings.<\/li>\n
  2. Go to Connectors.<\/li>\n
  3. Click Add custom connector.<\/li>\n
  4. Enter Name: Flevy<\/code>, and Remote MCP server URL: https:\/\/flevy.com\/mcp<\/code>.<\/li>\n
  5. Click Add, then start a chat.<\/li>\n<\/ol>\n

    ChatGPT<\/h3>\n

    Requires Developer Mode, available on paid plans.<\/p>\n

      \n
    1. Open Settings, then Connectors.<\/li>\n
    2. Turn on Developer Mode under the advanced or connectors area.<\/li>\n
    3. Return to Connectors and choose Create connector.<\/li>\n
    4. Enter Name: Flevy<\/code>, and MCP server URL: https:\/\/flevy.com\/mcp<\/code>.<\/li>\n
    5. Enable the connector inside a chat when you want to use it.<\/li>\n<\/ol>\n

      Some ChatGPT connector features are limited to Business and Enterprise workspaces.<\/p>\n

      Google Gemini<\/h3>\n

      Consumer path, through Gemini Spark.<\/p>\n

        \n
      1. On your computer, go to gemini.google.com.<\/li>\n
      2. Click Settings and help, then Connected Apps.<\/li>\n
      3. Under custom apps, add a new app using the MCP server URL https:\/\/flevy.com\/mcp<\/code>.<\/li>\n
      4. Follow the prompts to authorize it.<\/li>\n<\/ol>\n

        Google currently restricts this to eligible personal accounts, so check Gemini’s help pages for availability. For organizations, Gemini Enterprise connects custom MCP servers through the Google Cloud console as an admin setup.<\/p>\n

        Microsoft Copilot<\/h3>\n

        Through Copilot Studio.<\/p>\n

          \n
        1. Open Copilot Studio.<\/li>\n
        2. Go to your agent, then Tools or Add a tool.<\/li>\n
        3. Choose Model Context Protocol, or use the MCP onboarding wizard.<\/li>\n
        4. Enter the server URL https:\/\/flevy.com\/mcp<\/code> and complete the connection.<\/li>\n<\/ol>\n

          This is an agent-builder flow aimed at makers and admins rather than a one-click toggle in consumer Copilot chat.<\/p>\n

          Any Other MCP Client<\/h3>\n

          Point it at the endpoint https:\/\/flevy.com\/mcp<\/code> over Streamable HTTP.<\/p>\n

          How to Use the Flevy MCP<\/h2>\n

          Once Flevy is added to your AI, using it is simple. Just instruct the assistant to “use the Flevy MCP.”<\/p>\n

          The example below shows the same business question answered twice in a single chat. First on its own, then with Flevy.<\/p>\n

          The initial prompt, without using Flevy:<\/strong><\/p>\n

          \n

          I’m advising a mid-market industrial manufacturer that just acquired a smaller competitor. Walk me through what the first 100 days of integration should look like.<\/p>\n<\/blockquote>\n

          \"\"<\/p>\n

          The response is a clean three-phase outline. Stabilize, diagnose, execute quick wins. It is correct, and it is the kind of answer a sharp MBA could produce in a couple of minutes. It tells you what to think about.<\/p>\n

          The follow-up prompt, using Flevy within the same chat:<\/strong><\/p>\n

          \n

          Now use the Flevy MCP to go deeper. Search Flevy for the best post-merger integration frameworks, playbooks, and synergy-tracking models, pull the top one or two with their author credentials and what’s inside each, and find a real PMI case study of a comparable company. Then use all of it to turn your answer above into a plan I could actually run with this client.<\/p>\n<\/blockquote>\n

          \"\"<\/p>\n

          The Flevy-assisted response went and got the source material. It retrieved a 5-phase PMI methodology with IMO governance and 12 functional workstreams, a 150+ SOP first-100-days toolkit, both authored by McKinsey and BCG-trained consultants, plus an aerospace PMI case study that delivered 20% cost savings, a 15% efficiency gain, and 90% of milestones on schedule. Every source document was cited with a link back to Flevy.<\/p>\n

          It then built all of it into a 7-page plan with a governance RACI, a day-by-day 100-day workstream table, a synergy-tracking model, and advisor watch-outs. Ready to hand to the client.<\/p>\n

          Same assistant, same reasoning, with the world’s most comprehensive consulting knowledge base behind it. The difference in depth and output is night and day.<\/p>\n

          Other Ways to Use It<\/h2>\n

          The post-merger integration example is one use case. The connector works the same way across every management discipline in the library. A few prompts worth trying:<\/p>\n