writing-video-scripts
Writes what is said in a video — the words a person speaks on camera, or a voiceover heard over the picture.
Skill by SupercmoHQ
Chalk Labs
Need help deploying writing-video-scripts?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/supercmohq-writing-video-scripts
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: writing
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/writing-video-scripts
What's inside
- Skill name
writing-video-scripts- Description
- Writes what is said in a video — the words a person speaks on camera, or a voiceover heard over the picture. Fits the writing to the time available, so lines are never rushed to fit or padded to fill. A video longer than a single clip gets its script split into one piece per clip, each sized to that clip's seconds. Outputs the words as text only — nothing is recorded or voiced here. Use when the user wants a script, dialogue, or a voiceover written for a video.
- License
- Apache-2.0
- Metadata
- summary: Drafts spoken scripts built around proven opening hooks designed specifically to stop the scroll. It sizes the text to fit your exact runtime and paces words across natural breath points for human delivery. · version: 0.3.0 · category: creative
- Instructions
- 103 lines in SKILL.md · view on GitHub
Files (2)
- SKILL.md
- eval_cases.json · evals/
SKILL.md (first lines)
# Video Script Write the whole monologue first, to the video's total length, then cut it into one segment per clip. ## Workflow ### Step 1: Collect what the video has already fixed These are decided before the script is written, and the writing has to fit inside them: - **a brief of what the words should communicate** — the script is written from this, cut to fit the duration where it won't all fit - how long the video runs in total - how many clips it is, and how many seconds each one runs - which beats of the story land in which clip — what happens on screen, not what gets said - whether the words are spoken on camera or carried as a voiceover - which claims may be made, if any Where any of these is missing, ask for it rather than assuming it. ### Step 2: Work out the budget Speech runs at a fixed pace, so the word count isn't a style choice. - **Aim for 2-3 words per second of clip.** That is the target — a 15-second clip wants about 30 to 40 words. - **Never go past three words per second.** That is the ceiling, and it is the number Step 6 tests against. A 15-second clip fails at 46 words. The gap between the two is deliberate: it is room for the pauses, the breath and the beats where the picture carries the moment. A segment written to the ceiling leaves none of that. Add the per-clip budgets up against the total length before drafting anything. Where a segment runs long, cut words. Never speed up the delivery, and never stretch the clip to fit the writing. ### Step 3: Write the hook The first three seconds decide whether the rest gets watched, so write this line before any other. - **It fits in three seconds** — six to ten words at speaking pace. The viewer has already decided by then, so a line that needs longer has arrived too late. **Pick one pattern and commit to it.** These are the ones that recur across short-form: - **Problem** — name a symptom the viewer is living with right now. - **Contrarian** — contradict something the category takes as given. - **Curiosity** — point at a specific thing they don't know and signal it's about to be answered. - **Result first** — show the outcome with no explanation attached to it yet. - **Social proof** — someone else's experience or number, often relayed rather than claimed. Only where real proof was supplied; otherwise this pattern is closed. - **Scenario drop-in** — start mid-situation, already recognisable, with no setup. - **Pattern interrupt** — an ordinary scene with one thing visibly wrong, played straight. - **Listicle** — state a defined scope up front, so the viewer knows what they are committing to. ### Step 4: Write the monologue **The hook from Step 3 is the first line.** Everything after it carries that same thought forward and finally closes the gap it opened. **It has to sound spoken, not read.** If a viewer can tell the words are being read, the whole premise goes. - Write the way people actually talk — contractions, sentence fragments, a thought that restarts, the occasional aside. Full, balanced sentences are the clearest sign of a script. - **Specific beats general, every time.** A detail a viewer could check carries more than a general one. - **Make a point once.** Once a point has been made it is spent. **Write only what can be said aloud.** Commas and full stops are what a reader turns into breath, so punctuate with those. No dashes or em dashes; they are not spoken reliably. No brackets, no stage directions, no speaker labels, no emoji. ### Step 5: Split it into segments One segment per clip, each covering the beats Step 1 gave that clip. The finished video should sound like one take cut into pieces, not several recordings played in order. 1. **Cut where the ear already hears a break** — the end of a sentence, a comma, the join between two clauses. A cut mid-phrase leaves one clip trailing off on a fragment and the next opening on one, and the join stops reading as continuous. Mid-sentence is fine; mid-thought is the point. Mid-phrase is not. 2. **Prefer fewer, longer segments.** Where two neighbouring sentences both fit inside one clip's budget, keep them together. Every seam is a chance to sound stitched, so make as few as the durations allow. 3. **Don't change the words at the seam.** Joined back in order, the segments have to reproduce the monologue exactly — nothing reworded, reordered, added or dropped to make a split land. If a split only works by rewriting, the monologue is wrong rather than the split: go back to Step 4. 4. **Only the first segment introduces anything.** Every later one picks up mid-thought — no greeting, no saying what the product is again. ### Step 6: Check every segment against its clip Take each segment beside the clip it belongs to and test it. A segment that fails either test gets rewritten, not shortened at the edges — then tested again. - **Does it fit the seconds?** Count the words against that clip's budget from Step 2. Over budget is a rewrite: the delivery is never sped up and the clip is never stretched to fit the writing. - **Does it match what is on screen?** The words have to belong to the beats that clip actually shows. A segment describing something the viewer isn't looking at reads as a voiceover pasted over the wrong footage, even when every line is good on its own. Cutting frees budget. Spend it on a beat that earns it, or let the clip run quieter. ### Step 7: Hand the segments back One segment per clip, in order, each tied to its clip. Music is never added unless the user asks for it. Where they do, say so when handing back — that request has to carry through into the clip prompts.
From SupercmoHQ/superCMO-skills, Apache-2.0.
More from SupercmoHQ/superCMO-skills
- 0.796score
writing-video-prompts
SupercmoHQ
Writes the motion prompt for a video clip — what moves, how the camera behaves, the physics and the audio, in the shape the chosen model wants.
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
adapting-formats
SupercmoHQ
Resizes an existing image ad into the other shapes it has to run in — reads the ad first, then regenerates it at each ratio with the original passed in as a…
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
analyzing-brand
SupercmoHQ
Analyzes a brand from its own website — its colours and fonts, the logo and imagery it uses, how it sounds, who it sells to, the claims it makes, and what it…
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
analyzing-own-ads
SupercmoHQ
Audits the ads a company is running itself — what's live now, what it has already stopped, and which have run longest.
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
analyzing-products
SupercmoHQ
Normalizes a product into generation-ready facts — from an e-commerce URL (a clean description plus curated product images) or from a photo alone (category…
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
cloning-video-ads
SupercmoHQ
Clones an existing video ad — reads a supplied reference ad, keeps its structure, pacing, shots, camera and hook, and rebuilds it as a new video with the…
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
generating-ad-videos
SupercmoHQ
Makes a product ad video — a product showcase with no one on screen, or a story or lifestyle commercial where a presenter or actor plays a role in the brand's…
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
generating-product-photos
SupercmoHQ
ALWAYS read this skill before generating a product photography — before writing any prompt or calling image_generate.
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
generating-ugc-videos
SupercmoHQ
Generates a UGC video — a creator or customer, on camera as themselves, sharing their own authentic take on a product.
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
generating-videos
SupercmoHQ
ALWAYS read this skill before generating or animating any video, or calling video_generate — text-to-video, image-to-video, a start→end transition, or a…
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
planning-campaigns
SupercmoHQ
Decides what to make next — turns the brand, the product, your own running ads and competitor research into the concepts worth testing next.
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
researching-competitor-ads
SupercmoHQ
Analyzes the ads competitors are running — the ones live now, the ones they have already stopped, and which have run longest.
SkillAFreshness: A★ 206Claude Code +9
Signals
- Stars
- 206
- Forks
- 23
- Open issues
- 9
- Contributors
- 4
- Last commit (this folder)
- Aug 13, 2026 (2 months ago)
- Last release
- Oct 21, 2026 (9 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- Version
- 0.3.0
- License
- Apache-2.0
- Repository
- github.com/SupercmoHQ/superCMO-skills
- Website
- getsupercmo.ai
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from SupercmoHQ
- 0.796score
adapting-formats
SupercmoHQ
Resizes an existing image ad into the other shapes it has to run in — reads the ad first, then regenerates it at each ratio with the original passed in as a…
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
analyzing-brand
SupercmoHQ
Analyzes a brand from its own website — its colours and fonts, the logo and imagery it uses, how it sounds, who it sells to, the claims it makes, and what it…
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
analyzing-own-ads
SupercmoHQ
Audits the ads a company is running itself — what's live now, what it has already stopped, and which have run longest.
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
analyzing-products
SupercmoHQ
Normalizes a product into generation-ready facts — from an e-commerce URL (a clean description plus curated product images) or from a photo alone (category…
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
cloning-video-ads
SupercmoHQ
Clones an existing video ad — reads a supplied reference ad, keeps its structure, pacing, shots, camera and hook, and rebuilds it as a new video with the…
SkillAFreshness: A★ 206Claude Code +9 - 0.796score
generating-ad-videos
SupercmoHQ
Makes a product ad video — a product showcase with no one on screen, or a story or lifestyle commercial where a presenter or actor plays a role in the brand's…
SkillAFreshness: A★ 206Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get writing-video-scripts running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
writing-video-scripts skill. Agent Mart. https://agentmart.dev/skills/supercmohq-writing-video-scripts (last updated 2026-10-24)