Skip to content

uplika/uplika-plugin / skills/hooks

Hook formulas (YouTube, Instagram, short-form) with Korean patterns; read by the writing skills, not called directly.

FileKindSize
SKILL.mdskill md3.0 KB
references/formulas.mdmarkdown35.2 KB

references/formulas.md

View on GitHub
# Hook formulas

Opening lines for the writing skills (`yt-script`, `threads`, `repurpose`, `schedule-week` and the channel skills). Pick a formula for the format, then fill it with the person's own specifics. Never fill it with a number, a result, a name or a date the person did not give you.

## Sources and licenses

- YouTube: adapted from Jakeschincariol/youtube-agent-skill, `skills/yt-script/hooks.json`. MIT License, Copyright (c) 2026 Jake Schincariol. Full text in `THIRD_PARTY_LICENSES/youtube-agent-skill.MIT`.
- Instagram: adapted from Jakeschincariol/instagram-agent-skill, `skills/ig-reel/hooks.json`. MIT License, Copyright (c) 2026 Jake Schincariol. Full text in `THIRD_PARTY_LICENSES/instagram-agent-skill.MIT`.
- Short-form: adapted from rediumvex/viral-hooks-skill, `hooks-database.md`. MIT License, Copyright (c) 2026 Roman Knox. Full text in `THIRD_PARTY_LICENSES/viral-hooks-skill.MIT`. That database names jakeolschewski/social-media-hooks-database (MIT) as its own starting point.
- Korean patterns: written for Uplika, not adapted from any of the above.

Changes from the originals: each formula is cut to one table row, the regular expressions the originals use to classify hooks are left out, the section introductions are shortened, and dashes used as pauses are replaced with commas, colons or full stops. Not affiliated with the original authors.

## YouTube

For the first seconds of a YouTube video. The opening has three jobs: confirm what the title promised, open a question the viewer cannot close yet, and show that the payoff exists. An opening that only does the last one is a summary.

| # | Formula | Shape | Example | Fails when |
|---|---|---|---|---|
| 1 | The Statistic | A number the viewer did not know, stated flat, then the consequence. | 97% of the people who start a channel quit before video 30. Here is what the other 3% do differently. | Dies if the number is vague or unverifiable. Say the source out loud. |
| 2 | Someone Else's Result | A named person's outcome, then the mechanism. | This channel went from 400 subscribers to 200,000 in seven months on one format. I mapped it. | Do not invent the number. If you cannot name the channel, use a different formula. |
| 3 | The Mistake | Name the error the viewer is probably making right now. | Your first 30 seconds are why nobody finishes your videos. It is not the topic. | Only works if the mistake is common and the viewer can check it in one second. |
| 4 | Contrarian Flip | State the accepted advice, then reject it. | Everyone says post daily. I posted once a week for a year and tripled the channel. | Needs evidence in the next 15 seconds or it reads as contrarian for its own sake. |
| 5 | The Reveal | Promise a specific thing will be shown, then show it. | I am going to show you the exact prompt, on screen, that wrote this video. | You have to actually show it, early. A reveal at 8:00 is a retention cliff at 1:00. |
| 6 | The Superlative | The single best/worst/fastest of a category. | This is the fastest way to turn one long video into thirty Shorts. | Superlatives are cheap. Earn it by naming what it beat. |
| 7 | The Clock | Bound the payoff in time. | In the next six minutes you will have a working title, thumbnail and script for your next video. | The clock has to be true. Overrun it and the next video's hook is not believed. |
| 8 | I Tried It | First-person experiment with a stated cost. | I let an AI run my channel for 30 days. I did not touch the uploads. | Needs a real cost, such as time, money or risk, or there is no stake. |
| 9 | The Question | Ask the exact question the viewer typed into search. | Why do your videos die at 30 seconds when the topic is fine? | Must be the viewer's question, not yours. Pull it from your own comments. |
| 10 | Before and After | Two states, one cut between them. | This was my thumbnail. This is my thumbnail now. Same video, four times the click-through. | Only works with a visual. Do not use it on a talking-head-only video. |
| 11 | The Teardown | Take a real thing apart in public. | I pulled the nine most-viewed AI videos of the month and broke down what the titles share. | Name what you pulled and how many. A teardown of one example is an anecdote. |
| 12 | The Stack | Two named tools combined into one outcome. | Claude plus your YouTube Studio export gives you next week's upload schedule in one prompt. | Both tools must be nameable and the outcome must need both. |
| 13 | The Warning | A cost the viewer is about to pay. | Do not upload another video until you check this one setting. | If the cost is small the hook is a lie. Reserve it. |
| 14 | The List | A counted set, with the count in the first line. | Seven things I would do differently if I started a channel in 2026. | The count must be exact and every item must be different in kind. |
| 15 | The Receipt | Show the artifact first, explain second. | This is the analytics page. This video did 40% of the channel's watch time. Here is why. | Requires a real screenshot. A described receipt is not a receipt. |
| 16 | The Insider | Information from inside a system. | YouTube tells you which videos are underperforming. Almost nobody opens the report. | Do not claim insider access you do not have. |
| 17 | The Impossible Claim | State something that sounds untrue, then prove it. | You can write, title, thumbnail and schedule a week of videos without opening a single editor. | The proof has to start within 20 seconds or it reads as clickbait. |
| 18 | The Comparison | Two named options, one winner. | I wrote the same video three ways. One of them held 62% and the other two did not. | Say the losing option out loud. A comparison with no loser is an advert. |
| 19 | The Origin | Where a result actually came from. | Every video on this channel starts in the same 12-line file. Here it is. | Works once per channel. It is a reveal about you, and you only have one. |
| 20 | The Deadline | Something is changing on a date. | The old description format stops mattering this month. Here is what replaces it. | Only use with a real, checkable date. This is the fastest formula to lose trust with. |
| 21 | The Direct Address | Name the exact viewer in the first six words. | If you have under a thousand subscribers, this is the only video you need this week. | Narrow beats broad. 'If you make videos' addresses nobody. |

## Instagram

For Reels, and for any short vertical video. A Reel has two hooks at once: the first spoken line and the first text on screen. Write them separately; the on-screen one is shorter and blunter, because it is read before it is heard.

| # | Formula | Template | Example | On screen | Best for | Trap |
|---|---|---|---|---|---|---|
| 1 | Cost Confession | {Specific amount} is what {one mistake} cost me. | $18,000 is what one missing clause cost me. | $18,000 MISTAKE | Trust, fast. People finish the videos where you are the one who looks bad. | A fake cost. 'It cost me my peace of mind' is not a number and everybody can hear that. |
| 2 | Negative Command | Stop {common action}. Do {alternative} instead. | Stop posting three times a day. Post twice and reply to every comment for an hour. | STOP POSTING DAILY | Saves and arguments. The imperative gets attention that a statement does not. | Telling people to stop doing something nobody does. The habit has to be one the viewer recognises in themselves. |
| 3 | Nobody Tells You | Nobody tells you that {uncomfortable truth about the thing they want}. | Nobody tells you that your first 30 reels are supposed to flop. | YOUR FIRST 30 WILL FLOP | New audiences. It positions you as the person who says the quiet part. | Picking a truth everybody already says. If the comments are agreeing before you finish, it was not a secret. |
| 4 | The Replacement | {This thing} replaced {expensive thing} for {price}. | This $12 tool replaced the editor I was paying $2,000 a month. | $2,000 -> $12 | Demos and tool videos. The price gap does the work. | Overstating the replacement. If it replaced 60% of the job, say 60%, or the comments will say it for you. |
| 5 | Time Collapse | This used to take me {long time}. It now takes {short time}. | Proposals used to take me five hours. They take twenty minutes now. | 5 HOURS -> 20 MIN | Process videos where the payoff is speed. Reads instantly on screen. | A ratio nobody believes. 40 hours to 4 seconds gets treated as a lie even when it is true. |
| 6 | The Receipt | I {did thing} for {N days}. Here are the actual numbers. | I posted one reel a day for 60 days. Here is what it did to my DMs. | 60 DAYS. REAL NUMBERS. | Experiments. Pairs with a screen recording of the real dashboard, which is the whole value. | Not showing the screen. A numbers video with no screen recording is a numbers claim. |
| 7 | Wrong Way, Right Way | You are doing {thing} wrong, and it is not your fault. {The fix}. | You are writing captions wrong, and it is not your fault. Instagram hides everything after line one. | YOUR CAPTIONS ARE CUT | Teaching. The 'not your fault' half stops it reading as an insult, which is what kills this one. | Skipping the absolution. 'You are doing it wrong' on its own makes people leave rather than learn. |
| 8 | Insider Leak | I spent {N years} {inside the thing}. Here is what we never said out loud. | I spent six years selling agency retainers. Here is what we never put in the proposal. | 6 YEARS INSIDE | Authority without a credential slide. Works only if the years are real. | Promising a secret and delivering the brochure. The comments are merciless about this one. |
| 9 | The Steal | Steal this {artifact}. It took me {time} to get right. | Steal this four-line follow-up. It took me two years and a lot of dead leads. | STEAL THIS | Saves and sends, which are the two metrics that move reach the most. | The artifact not being on screen. If they cannot screenshot it, there is nothing to steal. |
| 10 | If This, Then Watch | If you {specific situation}, this {short time} fixes it. | If your reels get views and no followers, the next 40 seconds fix it. | VIEWS BUT NO FOLLOWERS? | Qualifying hard. Fewer people stay, and the ones who do are the ones who buy. | A situation so broad it qualifies nobody. 'If you want to grow' is not a situation. |
| 11 | Numbered With A Favourite | {N} {things} that {outcome}. Number {k} is the one nobody uses. | Five caption edits that double saves. Number four is the one nobody does. | 5 EDITS. #4 IS THE ONE. | Retention through the middle: naming a favourite gives people a reason to stay past three. | N above seven. Nobody believes you have nine good ones, and you do not. |
| 12 | The Objection | "{Objection they actually say}." Fine. Here is the version that works anyway. | "I don't want to be on camera." Fine. Here is the faceless version that still sells. | "I HATE BEING ON CAMERA" | Selling to the people who already said no once. | Inventing the objection. Pull it verbatim from a DM or a comment, including the bad grammar. |
| 13 | Before And After, On Screen | {Show the old thing}. {Show the new thing}. {One sentence on the lever}. | This was my profile in March. This is it now. I changed the name field, nothing else. | MARCH / NOW | Anything visual. The cut between the two frames is the hook, the words are secondary. | Three levers. Name one thing that changed or the transformation stops being believable. |
| 14 | The Callout | {Specific group}, this is the one you have been ignoring. | Photographers charging under $2,000: this is the pricing page you have been avoiding. | PHOTOGRAPHERS UNDER $2K | Cutting a broad audience down to the one that converts. Send rate goes up because people tag each other. | A group so large it is not a callout. 'Entrepreneurs' is not a group. |
| 15 | The Flop Record | My first {N} {attempts} did nothing. Here is what changed on {N+1}. | My first 47 reels averaged 300 views. Number 48 did 400,000. One thing was different. | 47 FLOPS, THEN THIS | People at the start, which is most of the audience. High comment rate. | Making the change sound like luck. If the answer is 'I got lucky', there is no video. |
| 16 | The Verbatim Question | "{Question exactly as it was asked}". I get this every week. | "How do I post consistently without hating it?" I get this one every week. | ASKED EVERY WEEK | Search. A question typed the way people type it is what Instagram search is matching against. | Tidying the question up. The value is that it is in their words, typos and all. |
| 17 | Head To Head | {A} versus {B}. I ran both for {time} and one of them was not close. | Carousels versus reels. I ran both for 90 days and it was not close. | CAROUSELS vs REELS | Comments. People arrive already holding an opinion, which is the point. | Refusing to pick. A comparison with a diplomatic ending is a video with no ending. |
| 18 | Cold Open Demo | {Start mid-action, already doing the thing.} Watch what this does. | Watch what happens when I paste a competitor's caption into this. | WATCH THIS | Screen recordings and anything with a visible result. No setup at all, the hand is already moving. | A result that takes twelve seconds to arrive. If the payoff is slow, cut to it and show the wait afterwards. |
| 19 | The Deadline | {Thing} changes on {date}. Do {action} before then. | Your top nine grid crops change the second you post a reel. Fix the covers first. | BEFORE YOU POST AGAIN | Urgency that is real. Platform changes, seasonal windows, policy updates. | Manufactured urgency. One fake deadline and every future one is discounted. |
| 20 | Permission | You do not need {the thing everyone says you need}. | You do not need a niche. You need one person you can name. | YOU DON'T NEED A NICHE | Relief. Very high save rate, because it removes a blocker rather than adding a task. | Removing the blocker and putting nothing in its place. The second half is the whole video. |
| 21 | Mid-Sentence Start | {...and that is when {the turn}.} | ...and that is when the client asked for the money back. Three days after approving it. | HE ASKED FOR A REFUND | Stories. Starting mid-scene makes the viewer feel they arrived late, which is a reason to stay. | Never going back to fill in the gap. Deliver the setup by second eight or it reads as a mistake. |
| 22 | Contrarian Flip | {Well-known saying, inverted.} | Nice guys finish first. | NICE GUYS FINISH FIRST | Comments. The viewer has to stop and check whether they misheard you. | Flipping a saying nobody says. The original has to be a reflex, or the inversion is just a sentence. |
| 23 | The Statistic | {N}% of {group} {surprising fact}. | 33% of high school graduates never read another book. | 33% NEVER READ AGAIN | Sends. A number a person can repeat at dinner gets sent to somebody. | A statistic you cannot source. Say where it came from in the video or do not use it. |
| 24 | The Reveal | This is the new {thing}. | This is the new Sony point and shoot, and it fixes the one thing I hated. | THE NEW ONE | Anything physical or visual. The object is the hook and the words are captions on it. | Holding the thing up without saying what it changes. A reveal with no stake is an advert. |
| 25 | Someone Else's Result | There was one {person} who {outcome} in {time}. | There was one creator who grew a million subscribers in a single month. | 1M SUBS IN ONE MONTH | Proof when the user does not have their own number yet. Third-person is allowed if it is true. | No name and no link. An anonymous miracle is a rumour. |
| 26 | The Superlative | The {fastest / biggest / only} {thing} is {the answer}. | The fastest way to get confident is to build evidence you cannot argue with. | THE FASTEST WAY | One-claim videos with no story attached. It is the most common shape in the wild and the easiest to write badly. | A superlative with a soft answer. If the second half is 'consistency', you have said nothing and the comments will tell you. |

## Short-form

For text posts, Threads posts and chains, carousel covers, captions and short videos, grouped by the reaction each family relies on. Pair one formula that stops the scroll (negation, specificity, question) with one that keeps people reading (story, confession, emotional). The Platform column is the original author's tag; some of those platforms are not Uplika channels, and the shape still works on the channels that are. The examples are the original author's illustrations, not facts: never reuse their numbers or their health and money claims.

### 01 Curiosity Gap

Why it works: The brain treats unanswered questions like open browser tabs. It can't relax until the gap closes, so the viewer keeps watching.

| ID | Formula | Example | Platform |
|---|---|---|---|
| 001 | "Nobody talks about [unexpected thing] in [niche]" | "Nobody talks about why your Reels flop after 3am posts" | All |
| 002 | "I found out why [common problem] keeps happening" | "I found out why your followers never convert" | All |
| 003 | "The [niche] industry doesn't want you to know this" | "The skincare industry doesn't want you to know this" | TikTok / IG |
| 004 | "What happens when you [unexpected action] for [timeframe]" | "What happens when you post 3x a day for 90 days" | YouTube |
| 005 | "I've been [doing thing] wrong my entire life" | "I've been writing captions wrong my entire life" | All |
| 006 | "The truth about [topic] that nobody is saying" | "The truth about freelancing that nobody is saying" | All |
| 007 | "Why [common belief] is actually the opposite" | "Why 'eat less, move more' is actually making you fat" | All |
| 008 | "I wasn't supposed to share this, but…" | "I wasn't supposed to share this, but here's what big agencies charge" | TikTok / IG |
| 009 | "What I learned from [unusual source]" | "What I learned from a 70-year-old stripper about negotiation" | All |
| 010 | "The hidden cost of [popular thing]" | "The hidden cost of going remote that nobody mentions" | All |

### 02 Contrarian / Pattern-Interrupt

Why it works: Contradicting a belief creates cognitive dissonance. People engage to resolve it: agree or argue, both feed the algorithm.

| ID | Formula | Example | Platform |
|---|---|---|---|
| 011 | "Stop [common advice everyone gives]" | "Stop posting every day. Here's what actually grows" | All |
| 012 | "[Popular strategy] is dead. Here's what replaced it" | "Hashtag research is dead. Here's what replaced it" | All |
| 013 | "Unpopular opinion: [bold claim about niche]" | "Unpopular opinion: Canva templates are killing your brand" | LinkedIn / X |
| 014 | "I quit [thing everyone does] and [positive result]" | "I quit posting Reels and my engagement doubled" | All |
| 015 | "Everyone's doing [trend] but they're missing [key detail]" | "Everyone's doing carousels but missing the last slide" | IG / LinkedIn |
| 016 | "[Common advice] is making things worse" | "More cardio is making your body fat worse" | All |
| 017 | "Hot take: [contrarian belief]" | "Hot take: most personal brands are just glorified resumes" | X / LinkedIn |
| 018 | "I'm done with [popular practice]" | "I'm done with growth hacking. Here's what I do now" | All |
| 019 | "If you're still doing [X], stop" | "If you're still cold-emailing the same way, stop" | All |
| 020 | "The opposite of [common advice] worked for me" | "The opposite of 'hustle' grew my business 5x" | All |

### 03 Authority & Proof

Why it works: Specific numbers and proof activate the credibility filter. Vague claims get ignored. Precise claims earn attention.

| ID | Formula | Example | Platform |
|---|---|---|---|
| 021 | "I [impressive result] in [short timeframe]. Here's the breakdown" | "I grew to 100K followers in 6 months. Here's the breakdown" | All |
| 022 | "After analyzing [N] [things], here's what works" | "After analyzing 500 viral posts, here's what works" | All |
| 023 | "[N] years of [experience] taught me one lesson" | "8 years of freelancing taught me one brutal lesson" | LinkedIn |
| 024 | "I've helped [N] [people] do [result]. The #1 mistake is…" | "I've helped 200 creators monetize. The #1 mistake is…" | All |
| 025 | "My [metric] went from [X] to [Y] when I changed [one thing]" | "My CTR went from 2% to 11% when I changed thumbnails" | YouTube |
| 026 | "I tried [thing] for [N] days. Here's what happened" | "I tried cold showers for 30 days. Here's what happened" | YouTube |
| 027 | "[N] of my [content/clients] all do this one thing" | "All 12 of my 6-figure clients do this one thing" | All |
| 028 | "After [N] failures, I finally figured out [X]" | "After 47 failed launches, I finally figured out pricing" | LinkedIn |
| 029 | "Here's the data nobody is sharing about [topic]" | "Here's the data nobody is sharing about LinkedIn reach" | LinkedIn / X |
| 030 | "I tested [N] different [things]. Only one worked" | "I tested 13 different cold-email subject lines. One worked" | All |

### 04 Emotional Trigger

Why it works: Content that mirrors internal dialogue feels personal. The viewer thinks "this was made for me." That's the most powerful hook of all.

| ID | Formula | Example | Platform |
|---|---|---|---|
| 031 | "If you're struggling with [pain point], read this" | "If you're struggling to get past 1K followers, read this" | All |
| 032 | "This is for the [type of person] who [relatable struggle]" | "This is for the creator who posts daily and gets crickets" | All |
| 033 | "I almost gave up on [thing] until [turning point]" | "I almost gave up on YouTube until I found this" | YouTube |
| 034 | "The hard truth about [aspirational thing] nobody tells you" | "The hard truth about going viral nobody tells you" | All |
| 035 | "You're not [negative self-belief], you just need [solution]" | "You're not bad at marketing, you just need better hooks" | All |
| 036 | "If this hits home, you're not alone" | "If this hits home, you're not the only founder feeling lost" | All |
| 037 | "I cried writing this…" | "I cried writing this, but it needs to be said" | LinkedIn / IG |
| 038 | "To the person who needs to hear this…" | "To the founder who's about to quit: read this first" | All |
| 039 | "I see you. Here's what I wish I knew" | "I see you, struggling first-year freelancer. Here's what I wish I knew" | All |
| 040 | "If you've ever felt [emotion], watch this" | "If you've ever felt invisible at work, watch this" | TikTok / IG |

### 05 Listicle & Value Stack

Why it works: Numbers set expectations. Viewers calculate value-to-time before committing. "Save this" / "bookmark" = micro-CTA boosting engagement signal.

| ID | Formula | Example | Platform |
|---|---|---|---|
| 041 | "[N] [things] that will [outcome] (save this)" | "7 free tools that will 10x your content (save this)" | All |
| 042 | "The only [N] [things] you need to [achieve result]" | "The only 3 apps you need to run a faceless YouTube channel" | All |
| 043 | "I spent [time/money] on [thing] so you don't have to" | "I spent $2,000 on YouTube courses so you don't have to" | YouTube / IG |
| 044 | "[Thing] cheat sheet (bookmark this)" | "Instagram algorithm cheat sheet (bookmark this)" | All |
| 045 | "[N] things I wish I knew at [age/stage]" | "7 things I wish I knew at 25 about money" | All |
| 046 | "[N] tools every [profession/role] needs" | "5 tools every solo founder needs in 2026" | All |
| 047 | "[N] habits that changed my life in [timeframe]" | "4 habits that changed my life in 90 days" | All |
| 048 | "[N] mistakes I made so you can skip them" | "6 mistakes I made building my agency so you can skip them" | LinkedIn |
| 049 | "The ultimate [topic] starter pack" | "The ultimate freelance writer starter pack" | X / IG |
| 050 | "[N] underrated [tools/habits/strategies]" | "5 underrated AI tools that actually save me hours" | All |

### 06 Question Hooks

Why it works: Direct address forces the brain to formulate an answer. The viewer enters dialogue mode, can't stay passive.

| ID | Formula | Example | Platform |
|---|---|---|---|
| 051 | "Why does [common annoyance] keep happening?" | "Why does my plant keep dying even though I water it?" | All |
| 052 | "What if [shocking premise]?" | "What if everything you know about productivity is wrong?" | All |
| 053 | "Have you ever wondered why [common thing]?" | "Have you ever wondered why airline food tastes so bad?" | YouTube |
| 054 | "Are you making this [niche] mistake?" | "Are you making this LinkedIn profile mistake?" | LinkedIn |
| 055 | "Is this you?" | "Is this you at 4pm every workday?" | TikTok / IG |
| 056 | "Can you guess [shocking number/fact]?" | "Can you guess how much I made my first year freelancing?" | All |
| 057 | "What would you do if [scenario]?" | "What would you do if your biggest client fired you tomorrow?" | LinkedIn |
| 058 | "How long until [inevitable outcome]?" | "How long until AI replaces your entry-level marketing job?" | LinkedIn / X |
| 059 | "Did your [authority figure] ever tell you about [X]?" | "Did your school ever tell you how 401(k) fees work?" | All |
| 060 | "Why is nobody talking about [X]?" | "Why is nobody talking about the 4-day work week data?" | All |

### 07 Story & Narrative

Why it works: Mid-story openings hijack the brain's narrative-completion drive. The viewer needs the resolution.

| ID | Formula | Example | Platform |
|---|---|---|---|
| 061 | "Last week, [unexpected thing] happened" | "Last week, a stranger sent me $10,000" | All |
| 062 | "3 years ago I was [low point]. Today [high point]" | "3 years ago I was sleeping in my car. Today I just bought my parents a house" | All |
| 063 | "I almost made the worst mistake of my [career/life]" | "I almost made the worst hiring mistake of my career" | LinkedIn |
| 064 | "Yesterday a stranger said something that…" | "Yesterday a stranger said something that changed how I see freelancing" | All |
| 065 | "I just got [shocking message/email]" | "I just got an email that made me question everything" | All |
| 066 | "The day I realized [insight]…" | "The day I realized I was the bottleneck in my own business" | LinkedIn |
| 067 | "I never thought I'd say this, but…" | "I never thought I'd say this, but I'm leaving Twitter" | All |
| 068 | "POV: you're [scenario]" | "POV: you're a 32yo woman trying to lose 5 pounds" | TikTok / IG |
| 069 | "When [character] said [quote], I knew…" | "When my mentor said 'you're avoiding the work,' I knew he was right" | LinkedIn / X |
| 070 | "Here's a story that will change how you see [topic]" | "Here's a story that will change how you see customer service" | All |

### 08 Negation / Myth Buster

Why it works: Two-word negations stop the scroll in 0.3 seconds. They contradict assumed reality and demand explanation.

| ID | Formula | Example | Platform |
|---|---|---|---|
| 071 | "Not [common belief]." | "Not arthritis." / "Not a heart attack." / "Not anxiety." | TikTok / IG |
| 072 | "[Common belief] is a lie." | "'Find your passion' is a lie." | All |
| 073 | "It's not [common cause], it's [real cause]" | "It's not lack of motivation, it's poor sleep" | All |
| 074 | "Forget what they told you about [topic]" | "Forget what they told you about cold pitching" | All |
| 075 | "[Popular thing] doesn't work. Here's why" | "Affirmations don't work. Here's why" | All |
| 076 | "Stop blaming [usual suspect]. The real issue is…" | "Stop blaming the algorithm. The real issue is your hook" | All |
| 077 | "There's no such thing as [popular concept]" | "There's no such thing as 'work-life balance'" | LinkedIn |
| 078 | "[Big claim from authority] is wrong" | "What your doctor said about cholesterol is wrong" | All |
| 079 | "Myth: [popular belief]. Reality: [contrarian truth]" | "Myth: drink 8 glasses of water. Reality: you don't need to" | All |
| 080 | "If you still believe [X], read this" | "If you still believe 'follow your dreams,' read this" | All |

### 09 Specificity & Data

Why it works: Specific numbers, times, and amounts feel real. Vague hooks feel made up. Specificity = credibility.

| ID | Formula | Example | Platform |
|---|---|---|---|
| 081 | "[Specific number]: that's how many [thing]" | "$847: that's what I waste on subscriptions monthly" | All |
| 082 | "[Specific time]: the exact moment [thing happens]" | "3:07am: the exact moment cortisol wakes you in peri" | All |
| 083 | "$[amount]: what [common thing] really costs" | "$23,000: what one bad hire really costs you" | LinkedIn |
| 084 | "[N]%: the share of people who [X]" | "73%: the share of LinkedIn posts that get under 100 views" | LinkedIn |
| 085 | "[Specific date]: when everything changed" | "March 14, 2024: the day I deleted Instagram" | All |
| 086 | "[Number]:[number]: the ratio that matters" | "3:1: the talk-to-listen ratio that closes deals" | LinkedIn |
| 087 | "[Lab number/metric]: the threshold nobody mentions" | "Ferritin: 70. The threshold nobody mentions" | All |
| 088 | "[Time of day]: the exact window for [X]" | "6-8am: the exact window cortisol wants you outside" | All |
| 089 | "[Specific dose/amount]: what actually works" | "30 grams: the protein floor at breakfast" | All |
| 090 | "[Age]: when [X] starts and nobody warns you" | "35: when perimenopause starts and nobody warns you" | All |

### 10 Personal Confession

Why it works: Vulnerability is rare and rewarded. Real confession from a real person beats polished expert content for trust + share rate.

| ID | Formula | Example | Platform |
|---|---|---|---|
| 091 | "I'm about to share something embarrassing" | "I'm about to share something embarrassing about my first year freelancing" | All |
| 092 | "I'll admit it: I [confession]" | "I'll admit it: I bought followers in 2019" | All |
| 093 | "Nobody talks about this part of [experience]" | "Nobody talks about the loneliness part of running a business" | LinkedIn / IG |
| 094 | "I failed publicly so you can succeed privately" | "I failed publicly so you can build your agency without my mistakes" | LinkedIn |
| 095 | "Here's what I lied about for years" | "Here's what I lied about on every podcast, until now" | All |
| 096 | "The thing I was most afraid to say" | "The thing I was most afraid to say about my industry" | LinkedIn |
| 097 | "I'm tired of pretending [X]" | "I'm tired of pretending hustle culture is healthy" | LinkedIn |
| 098 | "Real talk: [vulnerable admission]" | "Real talk: I made $0 for 14 months before my first $10K month" | All |
| 099 | "I never thought I'd post this" | "I never thought I'd post this, but my biggest deal taught me a brutal lesson" | LinkedIn |
| 100 | "Behind the scenes of [aspirational thing]" | "Behind the scenes of my 6-figure month, including the panic attack" | All |

## Korean patterns

Written for Uplika. These are not the lists above translated: a translated English hook reads as an advert in Korean. Use them when the person writes in Korean, fill the brackets with their own specifics, and keep the endings their voice file uses.

| # | Pattern | Example | Use for | Watch |
|---|---|---|---|---|
| 1 | 아직도 [낡은 방법] 하세요? | 아직도 엑셀로 재고 관리하세요? | Calling out a habit the reader recognises in themselves | The habit must be common in this audience, or it reads as an insult |
| 2 | [구체적인 대상]이라면 이것부터 보세요 | 이제 막 스마트스토어 연 분이라면 이것부터 보세요 | Narrowing to the reader who will act | 「직장인이라면」 is too broad to qualify anyone |
| 3 | [행동]했더니 생긴 일 | 한 달 동안 매일 아침 블로그 썼더니 생긴 일 | An experiment with a real cost in time or money | Say what actually happened; a vague 「놀라운 일」 is clickbait |
| 4 | 결론부터 말하면, [답] | 결론부터 말하면, 첫 달 광고는 끄는 게 맞았습니다 | Information posts and blog openings where the reader came for an answer | Give the real answer first, then the reasons |
| 5 | [A] vs [B], 직접 [기간] 써 보고 고른 쪽 | 노션 vs 구글 문서, 직접 석 달 써 보고 고른 쪽 | Comparisons that end with a pick | Name the loser and why; a tie is not an ending |
| 6 | [금액] 쓰고 나서야 알게 된 것 | 광고비 200만 원 쓰고 나서야 알게 된 것 | Trust through a cost the person paid | Only an amount the person really paid and is willing to say |
| 7 | [일] [기간] 하면서 한 번도 안 한 것 | 카페 7년 하면서 한 번도 안 한 것 | Insider experience without a credential | The years must be real |
| 8 | 모르면 손해 보는 [주제] | 모르면 손해 보는 연말정산 공제 하나 | Practical tips with a clear loss | The loss must be concrete; loss framing on every post reads as an advert |
| 9 | [검색창에 치는 질문 그대로]? | 스마트스토어 첫 달 매출 0원, 정상인가요? | Search-led posts and videos | Keep the reader's own wording; tidying it loses the point |
| 10 | [하나]만 바꿨는데 [결과] | 썸네일 글씨만 키웠는데 클릭이 늘었습니다 | One change and its effect | Name the one thing; several changes make it unbelievable |
| 11 | 솔직히 말하면, [인정하기 싫은 것] | 솔직히 말하면, 첫 제품은 제가 봐도 별로였습니다 | A confession that earns attention | The admission has to cost something; a disguised boast reads as fake |
| 12 | [긴 시간] 걸리던 [일], 이제 [짧은 시간]이면 끝납니다 | 반나절 걸리던 견적서, 이제 20분이면 끝납니다 | Process and tool posts where speed is the payoff | A believable gap; an extreme one is read as a lie |
| 13 | [날짜]부터 달라지는 [제도], 지금 할 일 | 다음 달부터 달라지는 배달앱 수수료, 지금 할 일 | A real, checkable change | Real dates only; one invented deadline discounts every later one |
| 14 | [통념]이라고요? 저는 반대로 했습니다 | 매일 올려야 큰다고요? 저는 일주일에 한 번만 올렸습니다 | Contrarian takes | The proof has to follow right away |
| 15 | ...그 말을 듣고 [행동] | ...그 말을 듣고 그날 바로 계약을 취소했습니다 | Stories that start in the middle of the scene | Fill in the setup soon after, or it reads as a mistake |
| 16 | [다들 필요하다는 것] 없어도 됩니다 | 블로그 시작하는 데 비싼 카메라 없어도 됩니다 | Relief: removing a blocker | Put something in its place; that second half is the post |
| 17 | [상황]이면 저장해 두세요 | 이사 앞둔 분이면 저장해 두세요 | Checklists and how-tos people come back to | The post has to be worth coming back to |
| 18 | 내돈내산 [제품] [기간] 써 본 후기 | 내돈내산 무선청소기 석 달 써 본 후기 | Reviews, especially on Naver Blog | Only when the person really paid; a sponsored post says so |
| 19 | [사소한 것] 하나 때문에 [손해] | 계약서 한 줄 때문에 보증금을 못 돌려받았습니다 | Mistakes with a single cause | The cause and the loss must be real |

Writing notes for Korean openings:

- Front-load the hook. Korean puts the verb last, so the words that carry the hook (the number, the reader, the keyword) go first.
- Address the reader by their situation (`이제 막 시작한 분`) rather than `당신`, which reads stiff or confrontational in most posts. Use `여러분` sparingly.
- Avoid translated phrasing such as `~에 대해 알아보겠습니다`, `이것이 바로 ~입니다` and `당신이 몰랐던 ~`. They read as translations or adverts.
- Cut intensifiers that promise without saying anything: `진짜`, `완전`, `대박`, `미쳤다`, `레전드`.
- Endings come from the voice file. Keep one register in a post (해요체, 합니다체 or 반말), never a mix.
- Write numbers the way people say and search them (`석 달`, `200만 원`), and only real ones.
- On Naver Blog the title follows the keyword set's `draftPrompt`; these patterns are for the opening lines of the body.
- On Threads, give the hook a line of its own and break the line after it.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 471kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall