Teach Through a Real Problem
Learn theory by solving a realistic problem step-by-step rather than reading passive textbook text.
Search by task, keyword or tag, then filter by category, AI tool or level.
33 prompts found
Learn theory by solving a realistic problem step-by-step rather than reading passive textbook text.
Why each step is needed, then a clearly marked final answer.
One hint at a time until you can finish the problem yourself.
Pinpoint the broken step, explain it, then show a correct path.
Compare how beginners vs experienced practitioners analyze signals, risks, and trade-offs.
Walk a messy incident or business miss through causes, evidence, and the one change that would stop a repeat.
Think mode: restate, step, check constraints, then verify the answer a different way than you derived it.
When a problem keeps returning: a full 5-Whys chain, then a six-category fishbone so you don't miss a cause the chain skipped.
Strip a tangled approach down to hard constraints vs habits, then rebuild 2 - 3 solutions that don't just copy the old one.
Map reinforcing and balancing loops behind a problem that keeps undoing the fix, then pick a high-leverage intervention.
Evaluates student work step-by-step to pinpoint the exact line where reasoning broke down.
Learn through realistic case studies with decision checkpoints, trade-offs, and consequences.
Practice real-world decision making under evolving constraints like memory, budget, and scale.
Run Five Whys on a problem that keeps coming back, and stop at a cause you can actually change.
Turn unranked priorities into weighted criteria, scored options, a sensitivity check, and a three-sentence recommendation.
Isolate a cause before you fix it: ranked differentials, cheapest tests first, a branched diagnostic path, then fixes with rollback.
Grok steels your argument, then attacks it and gives a real verdict - not a both-sides non-answer.
Search for real precedent, then argue the most likely failure. Stay adversarial until the verdict. DeeperSearch if you have it.
Twenty uncensored ideas across SCAMPER lenses, then clusters, scores, and three to actually pursue.
Assume the plan already failed, list specific failure mechanisms, and change the plan while failure is still cheap.
A move-specific SWOT, a PESTLE scan, and the cross-map that actually tells you go / no-go / go-with-conditions.
A yes/no call with one-time vs ongoing costs, certain vs speculative benefits, break-even, and a real counter-case.
Attack every assumption the plan depends on, rank the risky ones, and name the single point of failure - then a closing steelman.
Score competing work on Reach, Impact, Confidence and Effort, show the math, and sanity-check surprising ranks.
Separate positions from interests, spot false conflicts, and draft a path forward each side could actually accept.
Develop intuition by predicting cause-and-effect outcomes when system parameters change.
Understand a system architecture by starting with realistic failure modes and debugging.
Infer internal state and mechanics by observing system inputs, outputs, and behaviors.
Write a one-page decision memo: options, criteria, recommendation, and what would change your mind.
Generate ideas in quantity, then filter them with constraints so you leave with a shortlist, not a vibe.
Understand why a concept exists and what problem it solves before learning the technical details.
Generate a full project document - product, requirements, architecture, security, MVP, and launch plan - suitable for planning, review, or client approval.
Identify emotion and root cause, then write an empathetic reply with a permitted remedy - or ask clarifying questions when policy details are missing.