← Blog·Reference7 min read

Make Your AI Know You: Using Your Report with ChatGPT, Claude & Gemini

Generic AI gives generic career advice — because it has no idea who you are. Your report fixes that in five minutes. Here's the setup, tool by tool.

The problem: your assistant doesn't know you

Ask ChatGPT “should I start a YouTube channel or build a SaaS?” and you get the same balanced non-answer as everyone else. Not because the model is weak — because it's advising a stranger. It doesn't know that you score high on depth and low on visibility appetite, that your capacity is 10 hours a week, or that your market timing light is yellow.

Every major assistant now has a place to put exactly this kind of context. What most people are missing isn't the feature — it's a good who-am-I fileto put there. That's what your report is.

The file you already have

Every paid report ships in two formats: the PDF for you, and a Markdown file for your AI— attached to your report email, and downloadable via the “MD” button on your report page. Markdown is the format language models read best: clean headings, no layout noise. It contains your role profile, your strengths, your blind spot, your market timing, and your 12-week plan.

ChatGPT

1Projects (best option)

Create a project (e.g. “My career”), upload the .md file to its files, and every chat inside the project knows your profile. Ask anything career-related there.

2Custom Instructions (always-on)

Settings → Personalization → Custom Instructions. Under “What should ChatGPT know about you?”, paste the briefing section of your report (the 200-word summary at the top — the field has a character limit, so use the summary, not all 60 pages).

3Memory (incremental)

Or simply write in a chat: “Remember this about me:” followed by your top role and key strengths. ChatGPT stores it and applies it in future conversations.

OpenAI's official walkthrough of Custom Instructions.

Claude

1Create a Project

In claude.ai, create a project and upload your .md to the project knowledge. Every conversation in that project reads it automatically.

2One line of instructions

In the project instructions, add: “The attached roletype report describes who I am. Use it as context for all career and work advice.”

Anthropic's official intro to Projects.

Gemini

Gemini's equivalent is Saved Info(Settings → Personal context): paste the short version of your profile there. For deeper sessions, attach the .md file directly to a chat — Gemini reads files fine, it just doesn't persist them across chats the way Projects do.

Five prompts to try once it knows you

“Given my role profile, critique this plan: [your idea]. Where does it fight my profile instead of using it?”
“Rewrite my LinkedIn About section so it matches my top role and my actual strengths.”
“I'm in week 3 of my 12-week plan and stuck. Look at the plan and tell me what to cut.”
“My blind spot section says [X]. Design a weekly 15-minute routine that counteracts it.”
“Which of these two job offers fits my profile better — and what would each cost me?”

Pro corner

If you work with AI beyond the chat window, the same file goes further: coding agents like Claude Code or Codex read a CLAUDE.md or AGENTS.mdat the project root — a “who I am” section there shapes how they communicate and prioritize. Personal agent frameworks (OpenClaw and friends) keep a persona file like SOUL.md — reference your profile from it. And if you build with the API, the briefing section makes a solid block in your system prompt.

One note on privacy

Your report is a personality profile — treat it like one. Only paste it into accounts you control, and if you don't want it used for model training, check your assistant's data settings (all three let you opt out). Uploading it to a shared or work account means your employer's admins may see it.

More walkthroughs

No report yet? The assessment is free — the agent-ready export comes with the full report.

Take the Free Assessment →

See what's inside the report →

Bastian Brand

Bastian Brand, Ph.D. — author of The Hype Cycle Playbook, the framework behind the roletype assessment and this blog. About the author →