Hey, it's Cindy 🌱 You commented focus, so here is the complete setup from the reel: a tutor that tests you, a source-comparison prompt, and a conclusion-first response rule.
btw the waitlist for AI System Club is now open! If you're keen on using AI to automate your content, video editing, job search, business or build AI agents, feel free to sign up (only if you want to hehe). Ps. the waitlist gets first dibs:)
the outcome
What you get
A one-week learning sprint, a five-source synthesis, and a short answer format that preserves evidence without burying the conclusion.
Use the guide in order. Gather the real inputs, run the prompts against those inputs, inspect the evidence, then make one reviewable change at a time.
start here
Set it up step by step
Choose one skill or question that matters this week. Define a finished output, not a vague goal to learn more.
Build a seven-day curriculum with one daily lesson, one retrieval test, and one small applied task.
For videos, papers, or reports, supply the original material and run the source-comparison prompt. Do not ask the model to reconstruct a source from memory.
Set the conclusion-first rule for long answers, with evidence and uncertainty underneath.
Track completed tests and applied outputs. Time spent chatting is not the success metric.
copy these
The prompts
1. Turn Claude into a tutor
📋 copy this
Teach me [topic] through a seven-day sprint. Start with a five-question diagnostic and adapt the level from my answers. For each day, give me one concept, one worked example, three retrieval questions without answers, and one small task that produces something real. Do not reveal the answers until I attempt them. At the end of each day, grade my reasoning, explain the gap, and schedule one spaced-repetition question from earlier days.
2. Compare sources without flattening them
📋 copy this
I will give you up to five source files or links. For each source, state its main claim, evidence, method, date, incentives, and what it leaves out. Then build an agreement-and-disagreement table. Distinguish direct evidence from interpretation. Flag any claim that appears in only one source, and do not merge conflicting numbers. Finish with the three conclusions I can defend, the two uncertainties that remain, and the next primary source worth checking.
3. Make every answer conclusion-first
📋 copy this
For the rest of this session, answer in this order: one-sentence conclusion, three strongest reasons, evidence with source and date, uncertainty or counterargument, then optional detail. Keep the default under 250 words unless I ask for depth. Do not remove caveats that change the decision.
check the work
Quality gate before you trust it
1 The tutor tests recall before showing answers.
2 Every synthesis traces claims to supplied sources and keeps disagreements visible.
3 Conclusion-first answers retain uncertainty and source dates.
4 The week ends with a concrete artifact, decision, or demonstrated skill.
5 No health claim is made for a so-called dopamine detox; the guide treats it as an attention-design choice.
6 Save the source links, as-of date, and final evidence beside the output so another person can reproduce the check.
Guardrail. AI can compress retrieval, but summaries can omit context and tutors can still be wrong. Keep the original sources open, verify consequential claims, and do the recall and application work yourself.
keep it useful
Save the evidence with the workflow
Save the prompt with the source links and the date you verified them. When the tool, policy, price, interface, or documentation changes, update the guide before you reuse the workflow. Keep the inputs and outputs beside the project so the next run can be checked instead of reconstructed from memory.
the waitlist is now open
AI System Club is launching soon
Build AI systems with me through recorded lessons you can follow at your own pace. Join the waitlist for first access and the founding price.
btw this list gets first dibs the moment doors open :)