Skip to content
PromptifyLab

Learning & Education

Skills & Career prompts

10 prompts Free · no sign-up Works in ChatGPT, Claude & Gemini Search & filter these

Learning plans, practice design and honest feedback on where you are. Below are 10 copy-ready prompts. Fill in the [BRACKETS], copy, and paste into ChatGPT, Claude, Gemini or any capable assistant.

Learning a skill outside a course means having no curriculum and no feedback. AI supplies the first well and the second only if you force it to be honest.

The 10 prompts

Beginner 7 blanks to fill

Plan how to learn a new skill

Get from nothing to competent efficiently.

Prompt
Help me learn [SKILL].

WHY I WANT IT: [THE REASON AND WHAT YOU WILL DO WITH IT]
MY STARTING POINT: [WHAT YOU CAN DO NOW, AND ANY RELATED EXPERIENCE]
TIME AVAILABLE: [PER WEEK, AND FOR HOW LONG]
HOW GOOD I NEED TO BE: [able to do a specific task / competent / professional standard]
HOW I LEARN BEST: [IF YOU KNOW]
BUDGET: [FOR COURSES, TOOLS, EQUIPMENT]

Produce:

1. THE TARGET, MADE SPECIFIC - what you will be able to do, observably. 'Learn photography' is not a target; 'take usable product photos of my own stock' is. Rewrite my goal this way, because it determines what to learn and what to skip.

2. WHAT YOU ACTUALLY NEED - most skills have a large body of knowledge and a small subset that covers most real use. For my stated target, identify the subset. Learning the whole field when you need a specific capability is the most common way people spend six months and cannot do the thing.

3. THE SUB-SKILLS - the components, and which are the bottleneck. Progress usually stalls on one specific component, and identifying it early saves months.

4. THE SEQUENCE - what to learn first, and why. Order by dependency and by what unlocks practice: learn the minimum needed to start doing the thing badly, then improve, rather than studying until you feel ready.

5. THE FIRST PROJECT - something small and real to make or do in the first week. Learning attached to a concrete output progresses far faster than learning in the abstract, and it surfaces what you actually need to know.

6. THE PRACTICE STRUCTURE - what deliberate practice looks like for this skill: working at the edge of your ability, on a specific component, with feedback, rather than repeating what you can already do. Say what that means concretely here.

7. THE FEEDBACK PROBLEM - how you will know whether you are doing it right. This is the hardest part of self-directed learning and the most neglected. Options: a mentor, a community, a comparison standard, or a way to verify the result objectively. Say what is realistic for this skill.

8. THE PLATEAU - where progress will stall, roughly when, and what gets you past it. Most people stop at the plateau believing they have found their limit.

9. WHAT TO IGNORE - the parts of this field that are interesting and not needed for my target, and the equipment or tools that people buy and do not need.

10. THE REALISTIC TIMELINE - for my stated time and target. Honestly.

11. THE FIRST WEEK - specifically what to do, starting now.

What you get: A specific observable target, the subset actually needed, a dependency-ordered sequence, a first project and an honest feedback plan.

Tip: Point 5 is the shift that accelerates everything. Starting a real project badly in week one teaches you what to learn; studying until you feel ready does not.

Open in Written for Claude, ChatGPT, Gemini · Reviewed September 18, 2026
Intermediate 7 blanks to fill

Break through a plateau

Make progress when you have stopped improving.

Prompt
I have stopped improving at [SKILL]. Help me work out why.

WHERE I AM: [WHAT YOU CAN AND CANNOT DO]
HOW LONG I HAVE BEEN STUCK: [TIMEFRAME]
WHAT I PRACTISE: [DESCRIBE YOUR CURRENT PRACTICE HONESTLY]
HOW OFTEN: [FREQUENCY AND DURATION]
WHAT FEEDBACK I GET: [FROM WHOM, OR NONE]
WHERE I WANT TO GET TO: [THE NEXT LEVEL]

Produce:

1. THE DIAGNOSIS - plateaus have identifiable causes:
   - PRACTISING WHAT YOU CAN ALREADY DO: the most common by far. Comfortable practice is pleasant and produces no improvement.
   - NO FEEDBACK: repeating an error consistently embeds it. Practice without feedback can make you reliably worse.
   - A MISSING FUNDAMENTAL: a basic technique never properly learned, capping everything above it
   - WRONG PRACTICE STRUCTURE: whole-skill repetition rather than isolating the weak component
   - THE SKILL HAS CHANGED: what got you here is not what gets you further, and the next level requires something different in kind
   - NOT ACTUALLY STUCK: progress at intermediate level is slower and less visible than at the start
   From what I described, which applies.

2. THE COMFORT CHECK - in my described practice, how much is at the edge of my ability and how much is comfortable repetition. Be blunt about this.

3. THE WEAK COMPONENT - break the skill into parts and identify which is holding the rest back. Practising the whole skill when one component is the bottleneck produces almost no improvement in it.

4. THE FUNDAMENTAL CHECK - whether something basic was learned badly and is now the ceiling. Fixing this feels like going backwards and is usually the fastest route forward. It is also the thing people most resist.

5. THE FEEDBACK GAP - if I have none, this is probably the main cause. The realistic options for this skill, ranked by what is actually available to me.

6. THE DELIBERATE PRACTICE VERSION - what practice should look like: a specific component, at the edge of ability, with immediate feedback, repeated with adjustment. Describe it concretely for my skill.

7. THE MEASUREMENT - how to tell whether you are improving, since at this level the change is too slow to feel. Something recorded and compared over months.

8. THE DIFFERENT-IN-KIND QUESTION - whether the next level requires something categorically different from more practice: study, a teacher, a different environment, or working with people better than you.

9. THE HONEST POSSIBILITY - sometimes the plateau is where your current time investment leads, and going further requires substantially more time than you are giving it. Say if that is likely, so the choice is deliberate rather than a sense of failure.

10. THE ONE CHANGE - the single change most likely to produce movement.

What you get: A cause diagnosis, a comfort audit of your practice, the bottleneck component, a feedback plan and one specific change.

Tip: Point 2 is the uncomfortable one. Most practice at intermediate level is rehearsing what you are already good at, because it feels like progress and is not.

Open in Written for Claude, ChatGPT, Gemini · Reviewed September 18, 2026
Beginner 10 blanks to fill

Build a habit that actually holds

Make a behaviour stick beyond the first fortnight.

Prompt
Help me build a habit.

THE HABIT: [WHAT YOU WANT TO DO]
WHY: [THE REASON THAT MATTERS TO YOU]
HOW OFTEN: [DAILY / WEEKLY / SPECIFIC DAYS]
WHEN I HAVE TRIED BEFORE: [WHAT HAPPENED AND WHEN IT STOPPED]
MY CURRENT ROUTINE: [WHAT YOUR DAYS LOOK LIKE]
WHAT GETS IN THE WAY: [THE OBSTACLES]

Produce:

1. THE FAILURE ANALYSIS - I said what happened before. Habits fail in identifiable ways: too ambitious at the start, no specific trigger, relying on motivation, an all-or-nothing rule that collapses after one miss, or the habit not actually connecting to something I care about. Identify which happened.

2. THE SIZE PROBLEM - the most common cause. The version I described may be too large to sustain. Give a version small enough that I could do it on the worst day of the month. Starting smaller than feels worthwhile is what makes it survive, and it can grow later.

3. THE TRIGGER - a habit needs a specific cue, not an intention. 'After I [existing routine], I will [habit]' is the structure. From my described routine, identify the specific attachment point.

4. THE FRICTION - what stands between the trigger and the action. Remove steps: prepare things in advance, keep what you need visible and to hand, and eliminate any decision. Each decision is a chance to not do it.

5. THE OBSTACLES - I listed what gets in the way. For each: a specific plan. 'If [obstacle], then [what I do instead]', decided now rather than in the moment.

6. THE MISSED-DAY RULE - decided in advance. Missing once is normal and does not matter; missing twice is how habits end. The rule is: never miss twice. Say what the reduced version is for a day when the full version will not happen.

7. THE TRACKING - something simple and visible. Not an elaborate system, which becomes its own project and then a source of guilt.

8. THE MOTIVATION WARNING - motivation is unreliable and will not be there on most days. The design should assume it is absent. Anything that depends on feeling like it will fail.

9. THE FIRST TWO WEEKS - what to expect, and the specific point at which people usually stop.

10. THE REVIEW - when to check whether this is working and adjust, rather than either abandoning it or persisting with something that does not fit.

11. THE HONEST QUESTION - is this habit connected to something I actually want, or something I think I should want? Habits attached to obligation rather than a real reason are the ones that do not survive.

What you get: A failure analysis of previous attempts, a version small enough for a bad day, a specific trigger, friction removal and a never-miss-twice rule.

Tip: Point 6 is the rule that matters most. One missed day is nothing; the second missed day is where habits actually die, and deciding that in advance prevents it.

Open in Written for Claude, ChatGPT, Gemini · Reviewed September 18, 2026
Intermediate 6 blanks to fill

Learn from someone better than you

Get the most out of a mentor, coach or teacher.

Prompt
Help me learn from someone more experienced.

WHO THEY ARE: [THEIR ROLE AND EXPERTISE]
MY RELATIONSHIP TO THEM: [manager / colleague / formal mentor / teacher / someone who agreed to help]
WHAT I WANT TO LEARN: [THE SKILL OR KNOWLEDGE]
HOW MUCH ACCESS I HAVE: [frequency and format]
MY LEVEL: [WHERE YOU ARE]
WHAT I HAVE ASKED THEM SO FAR: [IF ANYTHING]

Produce:

1. WHAT THEY CAN GIVE YOU THAT NOTHING ELSE CAN - the things only an experienced person provides: judgement about what matters, feedback on your specific work, the decisions they made and why, and the things that are not written down. Information is available everywhere; this is not. Focus the access here.

2. THE TACIT KNOWLEDGE PROBLEM - experienced people cannot easily articulate what they know, because it has become automatic. Asking 'how do you do this?' produces a partial answer. Better: ask them to do it and narrate, or show them a specific case and ask what they notice. Give the specific questions that extract tacit knowledge.

3. THE QUESTIONS TO ASK - specific and answerable, not broad. 'What should I focus on?' produces a platitude. 'Here is what I produced, what would you have done differently and why?' produces something useful. Write ten questions suited to my situation.

4. THE BRING-SOMETHING RULE - the most valuable use of access is showing them your actual work and getting a reaction. Prepare something concrete for every interaction rather than arriving with general questions.

5. WHAT NOT TO ASK - things you could look up, questions that make them do your thinking, and vague requests for advice. Respecting their time is what keeps the access open.

6. THE OBSERVATION OPPORTUNITY - if I can watch them work, what to pay attention to: what they do first, what they skip, where they slow down, what they check, and what they ignore. The decisions are more informative than the actions.

7. THE FOLLOW-THROUGH - acting on advice and reporting back is what makes people keep investing in you. Advice given and visibly ignored ends the relationship quietly.

8. THE DISAGREEMENT - what to do when their advice does not fit your situation. Understand the reasoning behind it before discarding it; experienced people are often right for reasons they have not explained.

9. THE ACCESS BUDGET - given how much access I have, what to prioritise. Limited access should go to judgement and feedback, not to information.

10. WHAT TO OFFER - relationships last when they are not one-way. What someone at my level can realistically offer someone at theirs.

What you get: What only an experienced person can give, questions that extract tacit knowledge, a bring-something rule and an access budget.

Tip: Point 4 changes these relationships. Arriving with your actual work and a specific question gets you something useful; arriving with 'any advice?' gets you a platitude.

Open in Written for Claude, ChatGPT, Gemini · Reviewed September 18, 2026
Intermediate 6 blanks to fill

Decide what skill to learn next

Choose where to invest learning time.

Prompt
Help me decide what to learn next.

WHERE I AM NOW: [YOUR ROLE, SKILLS AND SITUATION]
WHERE I WANT TO BE: [YOUR GOAL, if you have one]
OPTIONS I AM CONSIDERING: [THE SKILLS YOU ARE WEIGHING]
TIME AVAILABLE: [REALISTICALLY]
WHAT I ENJOY: [WHAT YOU ARE DRAWN TO]
CONSTRAINTS: [MONEY, TIME, OBLIGATIONS]

Produce:

1. THE GOAL CHECK - if I gave a goal, does any of my option list actually serve it? People frequently learn the skill that is most visible or most discussed rather than the one their goal requires. If none of my options serve the goal, say so and name what would.

2. THE SKILL AUDIT - what I already have, and which options build on that versus starting from zero. A skill adjacent to your existing ones is learned faster and is more valuable in combination than a skill learned in isolation.

3. THE COMBINATION EFFECT - the most useful positions are usually the intersection of two or three skills rather than excellence in one. Given what I already have, which option creates the most valuable combination?

4. FOR EACH OPTION - an honest assessment:
   - What it takes to become useful at it, in time
   - How much of it can be learned in my available time
   - Whether the bar for 'useful' is low or high in this area
   - How durable it is: skills tied to a specific tool date faster than skills tied to a way of thinking
   - What it enables that I cannot currently do

5. THE ENJOYMENT FACTOR - taken seriously rather than dismissed. A skill you find interesting gets practised; one you think you should learn does not. This is a practical consideration, not a soft one.

6. THE DEPTH VERSUS BREADTH QUESTION - given where I am, whether I am better served going deeper in something I have or adding something new. Both are legitimate and they have different payoffs. Say which my situation suggests.

7. THE OPPORTUNITY COST - learning one means not learning the others. Name what I give up.

8. THE REVERSIBILITY - which of these can be tried for a month at low cost, to find out whether I like it before committing. For reversible choices, trying beats analysing.

9. THE RECOMMENDATION - one, with reasoning stated so I can disagree with it.

10. THE FIRST STEP - a small, concrete thing to do this week to test the choice.

What you get: A goal alignment check, a combination analysis against your existing skills, honest per-option assessments and a low-cost first step.

Tip: Point 3 is the underrated one. Being reasonably good at two things that rarely combine is usually worth more than being excellent at one common skill.

Open in Written for Claude, ChatGPT, Gemini · Reviewed September 18, 2026
Beginner 6 blanks to fill

Learn efficiently from courses and books

Get value from learning material instead of consuming it.

Prompt
Help me learn effectively from this material.

WHAT I AM WORKING THROUGH: [COURSE / BOOK / TUTORIAL SERIES / DOCUMENTATION]
SUBJECT: [WHAT IT COVERS]
WHY: [WHAT YOU WANT TO BE ABLE TO DO]
HOW I HAVE BEEN WORKING THROUGH IT: [YOUR CURRENT APPROACH]
HOW FAR IN: [PROGRESS]
WHAT IS NOT WORKING: [IF ANYTHING]

Produce:

1. THE CONSUMPTION PROBLEM - watching, reading and following along feel like learning and produce very little. The feeling of understanding while someone else explains is not the same as being able to do it. Assess my described approach against this honestly.

2. THE ACTIVE VERSION - what to do instead, specifically for this material:
   - Before a section: predict what it will say, or attempt the problem first
   - During: pause and explain it back before continuing
   - After: do something without the material open
   The last one is what produces capability. If I only change one thing, this is it.

3. THE TUTORIAL TRAP - following a tutorial step by step produces a finished result and almost no transferable skill, because every decision was made for you. The fix: do it again without the tutorial, then do something slightly different with the same techniques.

4. THE COMPLETION URGE - finishing a course is not the goal; being able to do the thing is. Say where in this material I should stop and practise rather than continuing, and which sections can be skipped given my stated purpose.

5. THE PROJECT ATTACHMENT - a real thing to build or do alongside the material, so every section has somewhere to be applied immediately. Suggest one appropriate to my subject and progress.

6. THE NOTE QUESTION - notes from a course are usually transcription and rarely revisited. What is worth writing down: things that surprised you, things you got wrong, and your own questions. Not the content, which the material already contains.

7. THE SPACING - revisiting earlier sections at intervals rather than moving continuously forward. Say when to look back.

8. WHAT IS NOT WORKING - I said what is going wrong. Diagnose: material too advanced, missing prerequisite, too passive, no application, or the material itself being poor. Some courses are badly made and the problem is not you.

9. THE STOPPING DECISION - whether to finish this material or move to something else. Sunk cost applies here too, and an unfinished course that taught you what you needed is a success.

10. THE CHECK - how to tell whether I have actually learned a section: do something from scratch, from memory, with no reference.

What you get: A consumption audit, active methods for this material, the tutorial trap named, a project to attach it to and a from-scratch check.

Tip: Point 3 is worth taking seriously. Completing a tutorial teaches you to follow that tutorial; redoing it without the instructions is where the learning is.

Open in Written for Claude, ChatGPT, Gemini · Reviewed September 18, 2026
Intermediate 6 blanks to fill

Get better at something you already do daily

Improve at work you have been doing for years.

Prompt
Help me get better at [SKILL I ALREADY DO].

HOW LONG I HAVE DONE IT: [TIME]
MY LEVEL: [HOW GOOD YOU THINK YOU ARE]
WHAT I DO DAILY: [THE ACTUAL WORK]
WHAT FEEDBACK I GET: [FROM WHOM AND HOW OFTEN, or none]
WHERE I THINK I AM WEAK: [YOUR VIEW]
WHAT BETTER WOULD LOOK LIKE: [YOUR TARGET]

Produce:

1. THE EXPERIENCE TRAP - years of doing something produces improvement for a while and then stops. Beyond that point, experience accumulates without capability increasing, because the work is repetition of what you can already do and feedback is absent or delayed. Assess whether my situation looks like this.

2. THE FEEDBACK AUDIT - the main reason daily work does not improve people. Feedback at work is usually about outcomes rather than technique, arrives too late to connect to the action, and is absent for most of what you do. What feedback do I actually get, and what am I doing with no feedback at all?

3. THE UNEXAMINED PARTS - the components of my daily work I have never deliberately thought about, because they became automatic early. These are where the largest gains usually are, and they are invisible precisely because they are automatic.

4. THE BREAKDOWN - the skill split into components, with an honest assessment of which are strong, which are adequate, and which have been adequate for years without improving.

5. THE DELIBERATE PRACTICE AT WORK - how to practise while doing real work, which is the only realistic option for most people:
   - Pick one component and attend to it deliberately for a period
   - Predict before acting, then check
   - Review afterwards: what happened, what you would do differently
   - Seek feedback specifically on that component rather than generally

6. THE COMPARISON - finding out what better looks like. Studying someone more skilled doing the same work, examining good examples of the output, or having your work reviewed by someone stronger. Without a reference standard, you cannot see your own gaps.

7. THE REVIEW HABIT - a short structured review of your own work at intervals, which is the closest thing to feedback available if none exists.

8. MY STATED WEAKNESS - assess it. People are often wrong about their own weak points, usually identifying something visible while a less visible one costs more.

9. THE ONE COMPONENT - the single element to work on first, and how to work on it inside my actual daily work.

10. THE MEASUREMENT - how to tell whether this is working over months, since improvement at this stage is slow and easily missed.

What you get: An experience-trap assessment, a feedback audit, the automatic components you have never examined and one element to work on within real work.

Tip: Point 3 is where the gains hide. The parts of your work that became automatic in year one have not improved since, and nobody has ever looked at them.

Open in Written for Claude, ChatGPT, Gemini · Reviewed September 18, 2026
Beginner 2 blanks to fill

Tailor your CV to a specific job description

Match your real experience to what the employer is asking for.

Prompt
Tailor my CV to this job.

JOB DESCRIPTION: [PASTE]
MY CURRENT CV: [PASTE]
ANYTHING NOT ON MY CV THAT IS RELEVANT: [EXTRA EXPERIENCE]

Do this:
1. REQUIREMENTS MAP - a table: requirement from the job | my matching evidence | strength (strong / partial / missing)
2. REWRITTEN SUMMARY - 3 lines aimed at this role
3. REWRITTEN BULLETS - for my most relevant roles, in the format: action verb + what I did + result (with a number where I gave one). Only use facts from my CV and extra notes - mark [ADD NUMBER] where a result would help but I did not provide one.
4. KEYWORDS - the job's key terms that should appear naturally (applicant tracking systems often scan for these)
5. GAPS - requirements I do not meet, and honest ways to address them in a cover letter
6. WHAT TO CUT - content that does not help for this role

Never invent jobs, titles, qualifications or results.

What you get: A requirements-to-evidence map, a targeted summary, rewritten result-focused bullets, keywords, gaps and cuts.

Tip: The [ADD NUMBER] markers are where your CV gets stronger. 'Reduced response time' is weak; 'reduced response time from 2 days to 4 hours' gets interviews.

Open in Written for ChatGPT, Claude, Gemini · Reviewed September 25, 2026
Beginner 4 blanks to fill

Prepare strong answers for a job interview

Practise likely interview questions with structured answers from your real experience.

Prompt
Help me prepare for a job interview.

ROLE AND COMPANY: [DETAILS]
JOB DESCRIPTION: [PASTE]
MY BACKGROUND: [SUMMARY OR CV]
INTERVIEW TYPE: [phone screen / competency / technical / panel / final]

Do this:
1. The 10 questions I am most likely to be asked for this role, including 2 difficult ones
2. For each, help me build an answer using the STAR method (Situation, Task, Action, Result) from MY experience - ask me for a story where you do not have one, rather than inventing it
3. My answer to "Tell me about yourself" in under 90 seconds
4. 5 smart questions to ask the interviewer
5. Then run a mock interview: ask me one question at a time, wait for my answer, and give feedback on structure, specifics and length before the next question.

What you get: Likely questions, STAR answer structures from your real experience, an intro pitch, questions to ask and a live mock interview.

Tip: The mock interview is the valuable part. Answer out loud, then paste what you said - your spoken answers are usually twice as long as you think.

Open in Written for ChatGPT, Claude, Gemini · Reviewed September 25, 2026
Intermediate 5 blanks to fill

Negotiate a job offer confidently

Plan what to ask for and write the negotiation message.

Prompt
Help me negotiate a job offer.

THE OFFER: [SALARY, BENEFITS, START DATE, OTHER TERMS]
MY SITUATION: [CURRENT PAY, OTHER OFFERS OR INTERVIEWS - true only]
MARKET INFORMATION I HAVE: [SALARY RANGES FROM JOB ADS, SURVEYS, PEOPLE I KNOW]
WHAT MATTERS MOST TO ME: [e.g. salary, remote days, title, training, start date]
MY WALK-AWAY POINT: [MINIMUM I WOULD ACCEPT]

Produce:
1. MY CASE - 3 reasons, based on value to the employer, that justify the request
2. THE ASK - a specific number or package, with a reason for that number
3. THE MESSAGE - a warm, professional email or call script that shows enthusiasm for the role and makes the request
4. RESPONSES - what to say if they say: "that's our best offer", "we'll think about it", "we can't move on salary"
5. NON-SALARY OPTIONS to ask for if salary is fixed

Rules: do not bluff about other offers. Keep the tone positive - the goal is a better deal and a good start with the employer.

What you get: A value-based case, a specific ask, a negotiation message and responses to common replies, plus non-salary options.

Tip: Anchor your number to evidence (market ranges, scope of the role). A number with a reason is much easier to agree to than a number alone.

Open in Written for ChatGPT, Claude, Gemini · Reviewed September 25, 2026

Where AI actually helps here

  • Building a learning path from where you are to a stated goal
  • Designing practice that targets a specific weakness
  • Being the interviewer, the client, the difficult colleague for rehearsal

Where it falls down

  • Assessing your actual ability from your description of it
  • Honest feedback unless explicitly instructed. The default is encouragement
  • Knowing what your field currently expects — requirements move

The mistake almost everyone makes: Accepting the encouragement

Ask for feedback and you get a compliment sandwich. Ask for the three things that would most hold this back, ranked, no positives and you get something usable. If a model has never told you that something you made is not good enough, you have not been asking properly.

Free tool: Output Grader

Runs in your browser. No sign-up, nothing uploaded.

Open the Output Grader →

Questions people ask


Can AI build me a learning plan?

A good structure, yes — sequence, milestones, what to practise. It cannot know how fast you learn or what your field currently demands, so treat timings as placeholders.


How do I get honest feedback from AI?

Remove the ownership and forbid balance: ‘a candidate submitted this — list only what would hold it back, ranked, no positives’. Framing it as someone else’s work removes most of the politeness.