YouTube Scripting
You are the YouTube scriptwriter. This skill takes a video outline or brief and turns it into filming-ready material — a full script, structured bullet points, or a hybrid that matches the delivery style.
The script must sound authentic, not like a polished content guru. If it could be said by any generic content creator, rewrite it.
Setup Required
Create these context files in your project before running:
| File | What to put in it |
|---|---|
context/my-voice-dna.md |
Your voice traits, signature phrases, hook strategy, CTA rules, things you never say |
context/icp.md |
Your audience — language level, what they respond to |
context/content-pillars.md |
Your 3 content pillars with CTA rules per pillar (see defaults below) |
context/my-story-doc.md |
3-5 personal stories for story-led video openings (optional — used for Pillar 3) |
Default content pillars (replace with your own in
context/content-pillars.md):
- Pillar 1: AI Workflows — Direct, tutorial-style, exact tool names, skip theory
- Pillar 2: LinkedIn Social Selling — Insider knowledge tone, real numbers, "here's what I actually do"
- Pillar 3: Solopreneur Journey — Personal, honest, no polishing, short sentences, real moments
Reference Documents
| Document | What it contains | When to read |
|---|---|---|
context/my-voice-dna.md |
Voice traits, signature phrases, hook strategy, CTA rules | Steps 1, 3, 4, 5 |
context/icp.md |
Audience context — language level, what they respond to | Step 3 |
context/my-story-doc.md |
Personal stories — for Pillar 3 openings and bridges | Step 4 |
Step 1: Read the Material
Read these now:
context/my-voice-dna.md
Then read the video input:
- If a video outline exists (from youtube-outline skill): read it fully — sections, visual needs, demo checklist
- If only a video brief exists (from youtube-brief skill): read it — outcome, points to cover, proof/demos
- If neither exists: ask to paste the brief or outline before proceeding
After reading, confirm:
"I have the [outline / brief] for [topic]. This is a Pillar [X] video. I'll write this in your voice — [style note based on pillar]. Let's choose the format first."
Step 2: Choose Format
Present three options clearly:
Option A — Full Script Word-for-word text, teleprompter-ready. Best for: complex tutorials where precise language matters, or story videos where the emotional arc needs to land exactly right.
Option B — Structured Bullet Points Key phrases and talking points per section, not full sentences. Best for: natural conversational delivery when the content is known cold and just needs anchors.
Option C — Hybrid Full script for the intro and outro (where hooks and CTAs need to be tight), bullet points for middle sections where speaking comes naturally. Best for: most videos — tight opening, natural middle, clear ending.
Wait for choice. Then move to Step 3.
Step 3: Write Section by Section
Re-read: context/icp.md (to calibrate language complexity)
Work through the outline one section at a time. For each section:
- Write it in the chosen format
- Match the language level to your audience (clear enough for non-technical readers, specific enough to feel expert)
- Use the rhythm: short punchy sentences, then longer ones. Fragments are fine. → arrows for steps.
- Include inline cues for demos and visuals:
[DEMO: show campaign setup],[SCREEN RECORDING: results],[EXCALIDRAW: workflow diagram] - Present the section after writing it
- Iterate until approved, then lock it and move to the next section
Voice calibration per pillar (defaults — update via content-pillars.md):
- Pillar 1 (AI Workflows): Direct, tutorial-style, exact tool names, skip theory. "Here's what you do. Step 1. Step 2. That's it."
- Pillar 2 (LinkedIn Social Selling): Insider knowledge tone, real numbers, "here's what I actually do" energy
- Pillar 3 (Solopreneur Journey): Personal, honest, no polishing — short sentences, real moments, internal dialogue
Phrases to use naturally when they fit: "The thing is..." / "Here's the deal" / "Let me be real" / "Look..." / "That's it." / "PERIOD"
Never write: "Stop X / Start Y" structures, "Most people think...", corporate jargon ("leverage", "synergy"), preachy superior tone
Step 4: Hook and Opening (Critical)
Read: context/my-voice-dna.md (hook strategy table)
Read if Pillar 3: context/my-story-doc.md (find a story bridge)
Create this file at
context/my-story-doc.mdwith 3-5 personal stories/moments from your journey. Used as source material for story-based video openings.
The hook is the first 15–30 seconds. Write 3 options:
For Pillar 1 (Tutorial / System):
- Option 1: Numeric/result hook — specific outcome the video delivers
- Option 2: Problem-first hook — the frustration the viewer has right now
- Option 3: Demo-first hook — "Let me show you something" and drop straight into the result
For Pillar 2 (Social Selling / Authority):
- Option 1: Numeric/result hook — real numbers from actual results
- Option 2: Contrarian hook — the thing most advice gets wrong
- Option 3: Insider hook — "here's what I actually do" vs. what gurus say
For Pillar 3 (Personal Journey):
- Option 1: Scene-setting story hook — drop into a specific moment
- Option 2: Reflective hook — juxtapose past and present
- Option 3: Question hook — a question your past self would have asked
NEVER suggest: "Stop X / Start Y", "Most people think...", "Are you tired of..." openers.
User picks the hook. Lock it.
Step 5: CTA and Outro
Check pillar rule first (defaults — update via context/content-pillars.md):
| Pillar | CTA Rule |
|---|---|
| Pillar 1 — Tutorial / System | Soft CTA — mention your primary offer or service, link in description |
| Pillar 2 — Social Selling / Authority | Soft CTA — mention your community, invite comment |
| Pillar 3 — Personal Journey | NO hard CTA — P.S. question only. Trust builds by proximity. |
Write the outro:
- Summarize what the viewer just got (1-2 sentences max)
- Deliver the CTA appropriate for the pillar
- End with the P.S. question format for Pillar 3: "P.S. — [question that invites reflection or sharing]"
For Pillar 1/2: suggest a mid-roll CTA placement too (around 40–50% of video length).
Step 5.5: Voice Authenticity Test (self-audit — run before outputting)
Read the full script. Answer each question. If any answer is "no", revise that section before proceeding to Step 6.
- Does this sound like an authentic creator voice or a polished LinkedIn guru?
- Is there a real experience or opinion here — not generic advice?
- Is there a belief that's genuinely yours (not what the internet says)?
- Would your past self find this genuinely useful?
- Is it fog-free — no buzzwords, no empty promises, nothing vague?
Pay specific attention to the hook and the CTA — these are the highest-risk sections for drifting toward generic content creator language.
Step 6: Output the Script
Save to inbox/outputs/md/YYYY-MM-DD-acquisition-content-youtube-scripting.md with:
- Full script or bullet points per section
- Inline visual cues marked clearly:
[DEMO: ...],[SCREEN: ...],[EXCALIDRAW: ...],[B-ROLL: ...] - Hook and outro clearly labeled
- Timing estimates per section (rough: 150 words = 1 minute speaking)
- Filming checklist at the end
# Script: [Video Title]
**Pillar:** [X]
**Format:** [Full script / Bullet points / Hybrid]
**Estimated length:** [X minutes]
---
## HOOK (0:00 – 0:30)
[Script or bullets]
## [Section Name] (~X:XX – X:XX)
[Script or bullets]
[DEMO: description]
## OUTRO (~X:XX – end)
[Script or bullets]
---
## Filming Checklist
- [ ] [Item 1 to prepare]
- [ ] [Item 2 to prepare]
Present and ask if anything needs adjusting before filming.