An AI Fluency Scorecard just leaked as a feature coming to Claude's settings, and it grades how good you actually are at using AI. Here's what it looks like, every one of the 11 habits broken down, and how to score yourself before Claude does.
Claude is about to start grading you, not your work, but on how good you actually are at using AI. The AI Fluency Scorecard was just spotted as a feature coming to Claude's settings: it scans your activity across Chat, Cowork, and Claude Code, scores you on 11 observable habits, and hands you a number out of 11 plus where to improve.
Here's roughly what it looks like:
Illustration based on the leaked feature (via TestingCatalog) plus Anthropic's public AI Fluency framework. Exact design, habit grouping, and scoring may differ in the final release.
The 11 habits group under three skills from Anthropic's AI Fluency framework, Delegation, Description, and Discernment. Below is every habit broken down: what it means, what scores well versus poorly, and exactly how to level it up, the part the video didn't have room for.
Delegation is everything you get right before you type, knowing the goal and handing the AI the right job with the right context.
1. You set a clear goal. You know what 'done' looks like before you prompt. Good: "Write a 200-word cold email that books a call." Weak: "write me an email." Level up: put the outcome and the win condition in your first sentence.
2. You hand off a right-sized task. Not so big it's vague, not so small it's busywork. Level up: if a task has more than ~3 real decisions inside it, break it into steps.
3. You give it the context it needs. Background, materials, constraints, who you are. Weak: expecting it to read your mind. Level up: paste the real source material instead of describing it.
4. You pick the right tool or mode. A quick question vs. extended thinking vs. Claude Code vs. a Project. Level up: use Projects for anything recurring, and turn on extended thinking for hard reasoning.
Description is how precisely you frame the request. This is where vague inputs create vague outputs, and where most people lose the most points.
5. You name the audience. Who is this for? Good: "for busy founders who skim." Level up: describe the reader's knowledge level and what they actually care about.
6. You specify the format. Bullets, a table, an email, 5 options, a word count. Level up: tell it the shape of the output before it writes.
7. You set the tone. Punchy, formal, warm, plain. Level up: name a tone and a thing to avoid, e.g. "confident, no corporate jargon."
8. You give an example of 'good.' The single biggest quality boost there is. Level up: paste one example of the style or output you want it to match.
Discernment is quality control, the habit that separates fluent users from people who paste whatever they're handed. Anthropic's own research found people iterate plenty but question the output far too little, which makes this the rarest skill and the fastest way to stand out.
9. You check the facts. You verify claims, numbers, and citations instead of trusting them. Level up: ask "what here should I double-check, and what are you least sure about?"
10. You catch weak logic. You notice when the reasoning is shaky or it dodged the hard part. Level up: ask it to argue the opposite case, then see which side holds.
11. You iterate. You treat the first draft as a draft, not the final answer. Level up: push back with "this is 70% there, here's what's off" instead of starting over.
The pattern almost everyone has
Most people are decent at Description and weak at Discernment. If you fix one thing this week, start questioning the output, it's the habit with the most upside on the scorecard and in real life.
You don't have to wait for the feature to ship. Paste this into Claude and have it grade your real habits against the same 11, with a score and a plan to improve.
Open a recent Claude chat where you did real work, then paste this:
Copy this prompt
You are going to grade my AI fluency honestly, like a tough coach. I'll tell you how I actually used AI on a recent task (or grade me on our conversation above). How I used it: [paste your prompts / describe what you asked and how, or say "use our chat above"] Score me out of 11 against these habits, grouped by skill. For EACH habit give a ✓ (strong), ~ (partial), or ✗ (missing), plus one sentence of evidence: DELEGATION 1. Set a clear goal before prompting 2. Handed off a right-sized task 3. Gave enough context and materials 4. Picked the right tool / mode DESCRIPTION 5. Named the audience 6. Specified the format 7. Set the tone 8. Gave an example of "good" DISCERNMENT 9. Checked the facts 10. Caught weak logic / pushed on the hard part 11. Iterated instead of accepting the first draft Then give me: - My total score out of 11, and which of the 3 skills is my weakest. - The 2 habits that would raise my score the fastest, with one concrete change for each. - A rewritten version of my original prompt that would have scored 11/11. Be honest, not flattering. If I'd score a 5, tell me it's a 5.
Once you know the 11 habits, you don't have to think about them every time. This skeleton bakes delegation, description, and discernment into one reusable prompt, fill it in for anything that matters:
Save this and reuse it for any real task:
Copy this prompt
GOAL: [what I'm trying to achieve and what 'done' looks like] <- delegation CONTEXT: [the background, materials, and constraints you need] <- delegation AUDIENCE: [who this is for] <- description FORMAT: [exactly how I want the output structured] <- description TONE: [the style / voice, and what to avoid] <- description EXAMPLE OF GOOD: [paste an example or describe one] <- description Now do the task. Then: - Flag anything you're unsure about or any fact I should verify. <- discernment - Give me the strongest counter-argument, or what could be wrong with this. <- discernment - Tell me one way this could be better if I gave you more.
The Only AI Masterclass You Need
If this guide helped, but you’re looking to go deeper, I got you!! My 30-Day Challenge takes you from saving AI tips you never use to actually building with AI, step-by-step.
I show you exactly how I automated two e-commerce brands, my social media, and most of my personal life, then hand you the agents, workflows & systems to do the same. I’m teaching you every single thing I know with one lesson and one build a day.
Join the AI Masterclass →© 2026 Mariah Brunner. All rights reserved.