All articles

Product

Use CareerArchitect Inside Claude, ChatGPT and Your Other AI Apps

12 October 2026  ·  4 min read

Our new MCP connector brings your CV, saved jobs, fit reports and optimisation into the AI assistant you already use — with secure sign-in and no surprise spending.

More and more of us already think out loud with an AI assistant — drafting emails in Claude, brainstorming in ChatGPT, coding in Cursor. Your job search shouldn't live in a separate tab you have to keep switching to.

So we built the CareerArchitect MCP connector. Connect it once, and your assistant can work directly with your CareerArchitect account: your real CV, your saved jobs, your reports and your optimisation journeys. Here's what that means and how to set it up.

What is MCP, in plain English?

MCP — the Model Context Protocol — is an open standard that lets AI assistants use tools from other services. Think of it as a secure plug:

  • Your assistant asks; CareerArchitect answers. When you ask about your CV or a job, the assistant calls the right CareerArchitect tool and gets back your actual data — not a guess.
  • It works across apps. Claude, ChatGPT, Claude Code, Cursor and any other app that supports remote MCP servers can connect.
  • You stay in control. You sign in to CareerArchitect yourself and choose exactly what the assistant is allowed to do.

What you can do from a chat

The connector exposes 20 tools covering most of what you do on the website. In practice, you just ask:

  • "Summarise my CV's three weakest spots." Your assistant reads your Master CV and its appraisal, then explains what to fix first.
  • "Find product owner roles in Stockholm." It searches the web for your saved job titles and cities, shows you a shortlist, and imports the one you pick from a careers-page link or pasted description.
  • "Should I apply to the Spotify role?" It runs the CV-vs-job fit report and walks you through the scores and the recommendation.
  • "Optimise my CV for this job." It shows you the optimisation stages you can run, starts the ones you choose, and brings back the before/after report for every bullet that changed.
  • "Tailor my summary for this role and save it." You can rework wording together in the chat, then save it onto the job — your employers, job titles and dates stay locked.
  • "What's on this week?" It lists your interviews, applications and activities from your Insights calendar.
  • "CatalystOne rejected me." It finds the job, asks you to confirm, then updates your pipeline.

Why this beats copy-and-paste: when you paste a CV into a generic chatbot, it only sees text. Through the connector, your assistant works with the same structured CV, scores and job data that power CareerArchitect — so its answers are about your career, and anything it saves lands back in your account in the right place.

Your credits stay in your hands

Letting an AI act on your account raises a fair question: can it spend your credits on its own? No. Anything that costs credits works in two steps:

  • First, a quote. Your assistant asks CareerArchitect what the action will cost and tells you, along with your balance.
  • Then, your yes. Only when you agree does the same request run and the credits get used.
  • Reading is free. Looking at your CV, jobs, reports and calendar never costs anything.
  • Same wallet as the website. Trial credits work, and you can top up with a credit pack — no membership required.

Secure sign-in, with permissions you choose

The connector uses standard OAuth, the same kind of sign-in you use for "Log in with Google". Under the hood:

  • Your password never leaves CareerArchitect. Your AI app receives a token, not your login details.
  • No API keys to copy. Your app registers itself automatically when you add the server URL.
  • Read-only by default. On the consent screen you can allow read access only, or tick one box to also allow imports and analysis.
  • Disconnect in one click. Account → Connected apps lists every app with access, and you can revoke any of them at any time.

How to connect (about two minutes)

The exact menus differ slightly between apps, and our MCP setup page has step-by-step guides for each one. The short version:

  • Copy the server URL from the MCP page: https://careerarchitect.coach/api/mcp
  • Add it as a connector in your AI app — for example, Settings → Connectors → Add custom connector in Claude.
  • Sign in to CareerArchitect in the window that opens, and choose your access level.
  • Start asking. Try "Show my saved jobs and their scores" to check everything is working.

One limitation to know: LinkedIn job links can't be fetched. Paste the job description instead, or use the company's own careers page link.

Try it today

The connector is free to set up and works from day one of your trial — every new account gets a 14-day free trial with 2,500 credits.

  • New to CareerArchitect? Start your free trial, upload your CV, then connect your assistant.
  • Already a member? Head to the MCP page, copy the server URL and connect in a couple of minutes.

Your job search, wherever you already work — with the same honest scores, careful optimisation and no-surprise credits you get on the website.

Written by

CareerArchitect Team

Practical advice on CVs, job searching, and interviews — written by people who understand how the hiring process actually works.

Get Started

Ready to architect your next move?

  • Personalized AI analysis in seconds
  • Identify high-probability career pivots
  • Master the recruiter's narrative