Loslegen
10 Opening Hooks for Your Short Video, Checked

10 Opening Hooks for Your Short Video, Checked

Paste one paragraph about your planned Short or Reel and get 10 cold-open lines, each with on-screen text, a first-frame note and the beat where the video keeps the promise. Checked for length, sound-off reading and a promise kept by a beat you described.
#Video
Rating
Weitere Bewertungen erforderlich
Sold
0
How to use
Herunterladen

Most opening lines of a Short take too long to say, need sound to make sense, or promise something the video never shows. Shorts Cold-Open Hook Set turns one paragraph about a planned short vertical video (what happens, the payoff and when it lands, audience, tone, language) into 10 different cold-open lines for the first seconds. Each has the spoken line, a short on-screen text version, what must be visible in the first frame, the promise it makes and the beat where the video keeps it. It plans before you film; it does not see analytics and does not predict views.

Phase 0 — Intake. Needs what the video is about, the payoff (the reveal or result and roughly when), and the language. Asks at most 3 questions, only if one is missing.

Phase 1 — Beat map. The video written as 4-7 timed beats with the payoff marked, plus audience, tone, length and a stated planning speaking rate (2.5 words per second, an estimate).

Phase 2 — The set. 10 hooks in at least five types (question, result-first, contrast, myth, number, direct address, scene drop), each with spoken line, on-screen text, first frame, promise and the beat that pays it off.

Phase 3 — Validation. Real word counts and estimated seconds, sound-off reading, promise-payoff match against the beats you described (no bait), a stranger test, claim safety, and your avoid list.

Phase 4 — Review lenses. One model applies four review lenses in turn: Scroll-Past Viewer, Payoff Auditor, Sound-Off Reader, Claims & Tone Scout. For each finding, the other three lenses each write a challenge, corroborate or no-challenge line. A verdict tags each finding CONFIRMED, CORROBORATED, DISPUTED or NO CHANGE; fixes are applied and a top 3 to test is named.

Sample output

Output from running this Skill on a sample problem (the input is a hand-written sample; the output is the Skill's unedited response):

Input ([HAND-WRITTEN FICTIONAL SAMPLE - invented for testing.]):

30-second Short for my cooking channel: I make a crispy fried egg using only a cold pan - the egg goes into an unheated non-stick pan with a spoon of oil, then I turn the heat to medium. At about 0:18 I lift the pan and show the edges are lacy and crispy while the yolk is still runny, and at 0:24 I cut into it. Audience: students and young people who cook for themselves. Tone: relaxed, a bit playful. Voiceover in English, plus on-screen text. Please avoid the words "hack" and "life-changing".

**Summary:** Crispy fried egg from a cold pan. Payoffs at 0:18 (lacy, crispy edges and a runny yolk) and 0:24 (cut into the egg). English, 30 seconds, 10 hooks, SFW.

## Essentials

- **Topic / what happens:** A crispy fried egg made with only a cold pan. The egg goes into an unheated non-stick pan with a spoon of oil, then the heat goes to medium. (given)
- **Payoff:** At about 0:18 the pan is lifted to show lacy, crispy edges and a still-runny yolk. At 0:24 the egg is cut. (given)
- **Language:** English (given)
- **Platform:** YouTube Short, vertical (given)
- **Audience:** students and young people who cook for themselves (given)
- **Tone:** relaxed, a bit playful (given)
- **Length:** 30 seconds (given)
- **Voice:** English voiceover plus on-screen text (given)
- **Avoid list:** "hack", "life-changing" (given)

## Beat map

- `0:00`: Unheated non-stick pan on the stove, burner off. A spoon of oil goes in. The oil-before-egg order is (assumed). The voiceover hook plays here.
- `0:04`: The egg goes into the unheated pan. Time (assumed).
- `0:08`: Heat turned to medium. Time (assumed).
- `0:12`: The egg cooks as the pan heats up. Time (assumed). What is visible is not described.
- `0:18` **[PAYOFF]**: Pan lifted. Lacy, crispy edges, yolk still runny. (given)
- `0:24` **[PAYOFF 2]**: Cut into the egg (given). Runny yolk visible inside (assumed from the 0:18 description). Runs to 0:30.

**Anchors**
- Audience: students and young people cooking for themselves (given)
- Tone: relaxed, a bit playful (given)
- Language and length: English, 30 seconds (given)
- Voice: voiceover plus on-screen text (given)
- Avoid: "hack", "life-changing" (given)
- Speaking rate: 2.5 words per second. This is a planning estimate, not a measurement; time the real take.

## Hook set

| # | Spoken line | On-screen text | Type | Frame 1 | Promise | Paid at | Words / est. s / text words | F1 | Mute | Pay | Clear | Claim | Flags |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 01 | Preheating isn't the only way to get a crispy egg. | Myth: crispy eggs need preheating | M | Unheated non-stick pan with a spoon of oil, burner off | A crispy egg without preheating the pan | 0:18 [PAYOFF] pan lifted, crispy edges | 10 / 4.0 s / 5 | PASS | PASS | PASS | PASS | PASS | 4a DISPUTED; 3b flag |
| 02* | Crispy edges and a runny yolk — from a cold pan? | Cold pan: lacy edges, runny yolk | Q | Unheated pan with oil, burner off | One egg with both crispy edges and a runny yolk, from a cold start | 0:18 [PAYOFF] edges and runny yolk shown | 10 / 4.0 s / 6 | PASS | PASS | PASS | PASS | PASS | 3a CORROBORATED (applied); 3b flag |
| 03* | This egg started in a cold pan — now watch the middle. | Wait for the cut | R | Finished egg lifted in the pan, lacy edges (from 0:18) | The inside of the egg shown when it is cut | 0:24 [PAYOFF 2] cut into the egg | 11 / 4.4 s / 4 | FLAG: open on 0:18 shot | PASS | PASS, WARN: long wait | PASS (fixed) | PASS | F1; WARN long wait; 2a NO CHANGE; 3b flag |
| 04* | Three steps, and the first one is: don't turn the heat on. | 3 steps: oil, egg, heat | N | Unheated pan, spoon of oil going in, burner off | A three-step order to copy, starting with the heat off | 0:00–0:08 oil, egg in, heat to medium | 12 / 4.8 s / 5 | PASS | PASS | PASS | PASS (fixed) | PASS | 3b flag |
| 05 | You can crack the egg in before the pan's even warm. | No waiting for the pan | D | Unheated pan with oil, burner off | No waiting: the egg goes in straight away | 0:04 egg into the unheated pan | 11 / 4.4 s / 5 | PASS | PASS | PASS | PASS | PASS | 3b flag |
| 06* | Lazy start, fancy-looking finish: nothing gets preheated. | Low effort, fancy-looking egg | C | Unheated pan with oil, burner off | A low-effort setup ends in an impressive-looking egg | 0:18 [PAYOFF] pan lifted | 7 / 2.8 s / 4 | PASS | PASS | PASS | PASS | PASS | 1a CORROBORATED (applied); 3b flag |
| 07 | Okay, egg's in, burner's still off — trust me. | Burner off. Egg in. | S | Egg landing in the unheated pan, burner off (from 0:04) | This odd-looking start turns out fine | 0:18 [PAYOFF] pan lifted | 8 / 3.2 s / 4 | FLAG: open on 0:04 shot | PASS | PASS | PASS | PASS | F1; 2b DISPUTED; 3b flag |
| 08 | Raw at the start, lacy and crispy when I lift it. | Raw white to crispy lace | C | Raw egg sitting in the unheated pan (from 0:04) | The raw white turns into crispy lace | 0:18 [PAYOFF] pan lifted | 11 / 4.4 s / 5 | FLAG: open on 0:04 shot | PASS | PASS | PASS | PASS | F1; 3b flag |
| 09 | One egg, one spoon of oil, one cold pan. | One egg, one spoon of oil | N | Unheated pan, oil going in, egg in hand (egg from 0:04) | The setup shown: one egg, a spoon of oil, an unheated pan | 0:00–0:04 oil in, egg in | 9 / 3.6 s / 6 | FLAG: egg must be in the 0:00 shot | PASS | PASS | PASS | PASS | F1; 1b NO CHANGE; 3b flag |
| 10 | Heat goes to medium now, with the egg already in. | Now the heat goes on | S | Hand turning the knob to medium, egg already in the pan (from 0:08) | What the egg does as the pan heats up under it | 0:12 egg cooks as the pan heats; result at 0:18 [PAYOFF] | 10 / 4.0 s / 5 | FLAG: open on 0:08 shot | PASS | PASS | PASS | PASS | F1; 3b flag |

## Validation

**1. Count and uniqueness: PASS**
- There are 10 rows.
- First words are all different: Preheating, Crispy, This, Three, You, Lazy, Okay, Raw, One, Heat.
- Promises are all different.
- No two on-screen texts share a core phrase.
- No two spoken lines describe the same moment with the same nouns.
- There is one question row [02].

**2. Length [LEN]: PASS** (final version)

Spoken lines:
- `01: Preheating|isn't|the|only|way|to|get|a|crispy|egg = 10 -> 4.0 s`
- `02: Crispy|edges|and|a|runny|yolk|from|a|cold|pan = 10 -> 4.0 s`
- `03: This|egg|started|in|a|cold|pan|now|watch|the|middle = 11 -> 4.4 s`
- `04: Three|steps|and|the|first|one|is|don't|turn|the|heat|on = 12 -> 4.8 s`
- `05: You|can|crack|the|egg|in|before|the|pan's|even|warm = 11 -> 4.4 s`
- `06: Lazy|start|fancy-looking|finish|nothing|gets|preheated = 7 -> 2.8 s`
- `07: Okay|egg's|in|burner's|still|off|trust|me = 8 -> 3.2 s`
- `08: Raw|at|the|start|lacy|and|crispy|when|I|lift|it = 11 -> 4.4 s`
- `09: One|egg|one|spoon|of|oil|one|cold|pan = 9 -> 3.6 s`
- `10: Heat|goes|to|medium|now|with|the|egg|already|in = 10 -> 4.0 s`

On-screen text:
- `01: Myth|crispy|eggs|need|preheating = 5`
- `02: Cold|pan|lacy|edges|runny|yolk = 6`
- `03: Wait|for|the|cut = 4`
- `04: 3|steps|oil|egg|heat = 5`
- `05: No|waiting|for|the|pan = 5`
- `06: Low|effort|fancy-looking|egg = 4`
- `07: Burner|off|Egg|in = 4`
- `08: Raw|white|to|crispy|lace = 5`
- `09: One|egg|one|spoon|of|oil = 6`
- `10: Now|the|heat|goes|on = 5`

**3. Sound-off [MUTE]: PASS on all rows**
- 01, 04, 05, 06: the text reads as a promise over the cold pan.
- 02: the text now names the cold pan (see 3a).
- 03: "Wait for the cut" over the finished egg.
- 07, 08, 10: the text matches the action in frame.
- 09: the items listed are visible in frame.

**4. Frame 1 vs beat map [F1]**
- 01, 02, 04, 05, 06: PASS (0:00 beat).
- 03: Frame 1 comes from 0:18. The edit must open on that shot as a flash-forward, then cut to 0:00.
- 07, 08: Frame 1 comes from 0:04. The edit must open on the egg-in shot, or move it to 0:00.
- 09: The egg comes from 0:04. It must be in shot at 0:00.
- 10: Frame 1 comes from 0:08. The edit must open on the knob turning with the egg already in.
- In Notes for the Top 3 rows affected [08].

**5. Promise-payoff [PAY]: PASS on all rows**
- Rows 01, 02, 06, 07, 08 are paid at 0:18. 0:18 is exactly 60% of 30 seconds, so there is no long-wait warning.
- 03 is paid at 0:24 (0:24 > 0:18): WARN, long wait.
- 04 and 09 are paid at 0:00–0:08 and 0:00–0:04.
- 05 is paid at 0:04.
- 10 is paid at 0:12, with the result at 0:18.

**6. Stranger test [CLEAR]**
- In the draft, 03 ("started cold") and 04 ("it") were WARN. Both were fixed (see the change list).
- Final: PASS on all rows.

**7. Claim safety [CLAIM]: PASS on all rows**
- There are no health, money, safety, urgency or scarcity claims and no group-blaming.
- 01 is hedged ("isn't the only way") and shown on screen (see 4a).

**8. Avoid list and SFW: PASS**
- "hack" and "life-changing" appear in no row.
- All rows are SFW.

**Phase 3 change list:** two changes [03, 04]

- **03 spoken.** The stranger test flagged "started cold": it is unclear whether the egg or the pan is cold. "in a cold pan" was inserted. "what the middle does" was then trimmed to "the middle" to stay within 12 words.
  - Before: "This egg started cold — now watch what the middle does." (10 / 4.0 s)
  - After: "This egg started in a cold pan — now watch the middle." (11 / 4.4 s)
- **04 spoken.** The stranger test flagged "it" as an undefined pronoun. "it" was replaced with "the heat".
  - Before: "Three steps, and the first one is: don't turn it on." (11 / 4.4 s)
  - After: "Three steps, and the first one is: don't turn the heat on." (12 / 4.8 s)

**Counts** (final version)
- Rows: 10.
- Types used: 7 [Q, R, C, M, N, D, S]. Distribution: Q 1 [02], R 1 [03], C 2 [06, 08], M 1 [01], N 2 [04, 09], D 1 [05], S 2 [07, 10].
- Longest spoken line: 04, 12 words / 4.8 s.

## Review lenses

### Findings per lens

**Lens 1: Scroll-Past Viewer**
- **1a [06]:** Spoken line "Lazy start, fancy-looking finish — same egg, same pan." "same egg, same pan" gives a scroller nothing, and what is "lazy" is not clear on one hearing. Fix: change it to "Lazy start, fancy-looking finish: nothing gets preheated."
- **1b [09]:** "One egg, one spoon of oil, one cold pan." This sounds like the ingredient list of many recipe videos and creates little tension. No change proposed; keep it as the plain-list option for testing.

**Lens 2: Payoff Auditor**
- **2a [03]:** Paid at "0:24 [PAYOFF 2] cut into the egg". It comes after 60% of the video (WARN: long wait). Frame 1 "Finished egg lifted in the pan, lacy edges" also reveals the 0:18 look up front. No change proposed: the line names the one thing not yet shown ("watch the middle").
- **2b [07]:** "trust me". The promise "This odd-looking start turns out fine" is vague, so it is hard to tell whether 0:18 fully keeps it. Fix: "Okay, egg's in, burner's still off — watch it crisp up." (10 / 4.0 s)

**Lens 3: Sound-Off Reader**
- **3a [02]:** On-screen "Lacy edges and runny yolk" over Frame 1 "Unheated pan with oil". With the sound off, the cold-pan angle is lost and the text promises an ordinary fried egg. Fix: "Cold pan: lacy edges, runny yolk" (6 words).
- **3b [all rows]:** Every on-screen text could sit under bottom captions or buttons. This is a flag only; the lens cannot see the real layout. Place text in the upper or middle third.

**Lens 4: Claims & Tone Scout**
- **4a [01]:** On-screen "Myth: crispy eggs need preheating". Calling it a "myth" is an absolute claim. The video shows one egg in one non-stick pan. Fix: "Crispy egg, no preheating" (4 words).
- No avoid-list or SFW breach, and no tone clash, on any other row.

### Cross-check

- `[1a, 06] Lens 2: corroborate - "a vague line makes no checkable promise; the fix names the unheated pan the 0:00 beat shows" / Lens 3: no challenge / Lens 4: no challenge`
- `[1b, 09] Lens 2: no challenge / Lens 3: no challenge / Lens 4: no challenge`
- `[2a, 03] Lens 1: corroborate - "a scroller who sees the finished egg at 0:00 may feel they have already seen the result" / Lens 3: no challenge / Lens 4: no challenge`
- `[2b, 07] Lens 1: corroborate - "'trust me' gives a scroller no picture of what is coming" / Lens 3: no challenge / Lens 4: challenge - "'trust me' fits the stated relaxed, playful tone, and naming the crisp result would move this promise toward row 01's crispy-egg promise"`
- `[3a, 02] Lens 1: no challenge / Lens 2: corroborate - "the cold start is half of what the 0:18 beat pays; without it the text undersells the payoff" / Lens 4: no challenge`
- `[3b, all rows] Lens 1: no challenge / Lens 2: no challenge / Lens 4: no challenge`
- `[4a, 01] Lens 1: no challenge / Lens 2: challenge - "the 0:18 beat shows a crispy egg from an unheated pan, and one shown counter-example is enough to refute 'need'" / Lens 3: no challenge`

### Verdict

- **1a [06] CORROBORATED.** Raised by Lens 1 and corroborated by Lens 2 in the cross-check for 1a. Applied.
- **1b [09] NO CHANGE.** No lens proposed a change.
- **2a [03] NO CHANGE.** Corroborated by Lens 1 in the cross-check for 2a, but no lens proposed a change. The WARN long wait and the early reveal stay attached to row 03.
- **2b [07] DISPUTED.**
  - Lens 2, with Lens 1's corroboration in the cross-check for 2b, says "trust me" is too vague and proposes "watch it crisp up".
  - Lens 4 says "trust me" suits the stated tone, and naming the result moves the promise toward row 01's.
  - Row 07 stays as written. The user decides.
- **3a [02] CORROBORATED.** Raised by Lens 3 and corroborated by Lens 2 in the cross-check for 3a. Applied.
- **3b [all rows] NO CHANGE.** Flag only, moved to Notes.
- **4a [01] DISPUTED.**
  - Lens 4 says "Myth:" is absolute when only one egg is shown, and proposes "Crispy egg, no preheating".
  - Lens 2 says the shown result refutes "need", so the label holds.
  - Row 01 stays as written. The user decides.

**Totals:** 7 findings. CORROBORATED 2 [1a, 3a], DISPUTED 2 [2b, 4a], NO CHANGE 3 [1b, 2a, 3b], CONFIRMED 0.

### Changes applied

- **1a, row 06 spoken line:**
  - Before: "Lazy start, fancy-looking finish — same egg, same pan." (8 / 3.2 s)
  - After: "Lazy start, fancy-looking finish: nothing gets preheated." (7 / 2.8 s)
  - Phase 3 re-run: uniqueness PASS (first word Lazy, promise unchanged, nouns start and finish), LEN 7 -> 2.8 s PASS, MUTE PASS, F1 PASS, PAY 0:18 PASS, CLEAR PASS, CLAIM PASS, avoid list and SFW PASS.
- **3a, row 02 on-screen text:**
  - Before: "Lacy edges and runny yolk" (5)
  - After: "Cold pan: lacy edges, runny yolk" (6)
  - Phase 3 re-run: core phrase "cold pan" unique among on-screen texts, LEN 6 ≤ 7 PASS, MUTE PASS, F1 PASS, PAY 0:18 PASS, CLEAR PASS, CLAIM PASS, avoid list and SFW PASS.
- Rows: still 10. Counts line unchanged; longest spoken line is still 04 at 12 words / 4.8 s.

### Top 3 to test

- **02 (Q):** Names both textures the 0:18 lift shows. It pays off at the 60% mark with no long wait, and Frame 1 is already the 0:00 shot.
- **05 (D):** Speaks directly to someone cooking for themselves. Its concrete benefit has the earliest payoff, at 0:04.
- **08 (C):** A before and after the viewer watches happen. The edit needs to open on the 0:04 shot.

**Recommendation**

Picks: [02, 05, 08]
- 02: 3a CORROBORATED (applied); 3b flag (see Notes)
- 05: 3b flag (see Notes)
- 08: F1 flag, open on the 0:04 shot (see Notes); 3b flag (see Notes)

## Notes

**Assumptions to confirm**
- Beat times 0:04, 0:08 and 0:12 are assumed.
- Oil going in before the egg is assumed. This affects rows 04 and 09.
- The burner being off at 0:00 is read from "unheated".
- The 0:24 cut showing a runny yolk is assumed from the 0:18 description.
- Row 09 says "one egg, one spoon of oil". If you also add salt or anything else on camera, that line still holds, but the on-screen text should not suggest that is the full recipe.

**F1 for the Top 3**
- Row 08 must open on the 0:04 shot of the raw egg in the unheated pan, or that shot must be moved to 0:00.

**3b [all rows]**
- Keep on-screen text out of the bottom area where captions and buttons usually sit.

**Disputed items for you to decide**
- 2b [07]: "trust me" vs "watch it crisp up".
- 4a [01]: "Myth: crispy eggs need preheating" vs "Crispy egg, no preheating".

**Kept as test options**
- 1b [09] is the plain-list opener.
- 2a [03] keeps its long-wait warning.

**Claims**
- The rows describe what this one egg shows. They make no general cooking-science claims.

**How to test**
1. Film the Top 3 openings on the same body of the video.
2. Post them or A/B them the way your platform allows.
3. Compare your own retention over the first seconds.

**Honest limit:** These lines are reasoned from the words only. Nothing here predicts views or retention, and the seconds are estimates, so time the real takes.

**Offer:** I can write 10 more hooks of one type, hooks for your next video, or a rewrite of this set for a text-only version.

Use cases

  • Writing several opening lines for one Short before you film, so you can test them
  • Making sure the on-screen text works with the sound off
  • Checking each hook's promise against the beats you described, so none is bait
  • Cutting slow openings down to a line you can say in a few seconds

FAQ

Will these hooks get more views? It cannot say. It sees no analytics or trends and predicts nothing; the only real test is posting and reading your own retention.

Are the seconds exact? No. They are estimated from a fixed planning rate of 2.5 words per second; time your real take.

Does it edit the video? No. It writes text only.

Is it safe for work? Yes. SFW only, no fake urgency, no unbacked health or money claims.

Which tools? Written for SKILL.md agents: Claude Code, Codex CLI and OpenClaw. One-time download, no hosting, no API key, no network calls.

Not affiliated with or endorsed by YouTube, Google, Meta or TikTok; platform names are used only to describe what the plan is for.