agentleFS
puretechnycpuretechnyc/purebrain-skills/skills/content/script-to-speech/SKILL.md

script-to-speech

Skill0 starsChanged 6 months agoMITView on GitHub

10-point framework for optimizing written content for audio/voice delivery. Blog audio, training modules, video scripts, podcast content, voiceover copy.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every piece of content it suggests?

    line 140
  2. Distinctive:

    Before delivering any audio-destined script, verify:

    line 162

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Script-to-Speech Optimization

Purpose: Transform written content into audio-optimized scripts using a 10-point framework.

Apply to: blog audio, training modules, video scripts, podcast content, voiceover copy


The 10-Point Framework

Apply ALL 10 points to any content destined for voice/audio output.

1. Scene-Based Structure

Break content into distinct scenes rather than continuous prose. Each scene = one clear visual or idea.

WRONG: "Our platform helps marketing teams automate content creation, analyze
performance data, and collaborate more efficiently..."

RIGHT:
[SCENE: The overwhelmed marketer]
You have 47 tabs open. Three campaigns need copy. Your boss wants the report by noon.
[SCENE: The solution arrives]
What if one platform handled all of it?

2. Hook Openings

First 10 words must create tension, curiosity, or a direct question. No warm-up sentences.

WRONG: "Today we're going to talk about how AI can help your marketing team..."
RIGHT: "Most marketing teams are doing twice the work for half the results."
RIGHT: "What if your AI knew your brand better than your newest hire?"

3. PAUSE Markers

Insert [PAUSE] or [SHORT PAUSE] to control listener rhythm. Use after key points, before reveals, after questions.

The average marketing team spends 60% of their time on tasks AI can handle. [PAUSE]
Sixty. Percent. [SHORT PAUSE]
That's time you'll never get back.

4. Short Sentences (avg 12 words)

Target: 8-15 words per sentence. Ruthlessly cut long sentences in two.

WRONG: "The platform's advanced natural language processing capabilities enable
marketing teams to generate high-quality content at scale."
RIGHT: "The platform writes like your best copywriter. At ten times the speed."

5. Contractions Everywhere

Written English sounds stiff when read aloud. Contractions fix this.

WRONG: "You will not believe how much time you are saving."
RIGHT: "You won't believe how much time you're saving."

6. Active Voice

Passive constructions slow delivery and confuse listeners.

WRONG: "Content is generated by the AI based on your brand guidelines."
RIGHT: "The AI generates content using your brand guidelines."

7. Emotional Anchors

Plant 2-3 emotional moments: frustration, relief, excitement, pride. Listeners remember feelings.

[FRUSTRATION ANCHOR]
You've rewritten that headline seven times. It still isn't right.

[RELIEF ANCHOR]
Your AI suggests three options in seconds. All on-brand. All ready to test.

8. Reflection Closes

End sections or the full piece with a question or statement that lands in silence.

WRONG ending: "...and that's why this is the right choice for your team."
RIGHT ending: "The question isn't whether AI will transform your marketing.
It already has. The question is: are you using it?"

9. Rhythm Variation

Alternate long and short sentences to create natural spoken cadence.

Long: "Most marketing teams spend their days bouncing between tools, copying and
pasting, reformatting the same content for different platforms."
Short: "There's a better way."
Long: "One platform brings everything into one place..."
Short: "One login. One workflow."

10. Direct Address ("You")

Speak directly to the listener at least once every 3-4 sentences.

WRONG: "Marketing teams often struggle with consistency."
RIGHT: "Your team knows the struggle. The brand voice guide that nobody reads.
The off-brand post that went live anyway."

Before / After Example

BEFORE (Written prose): "Our platform is an AI-powered marketing tool designed to help teams streamline content creation, improve brand consistency, and measure campaign performance more effectively. The platform integrates with existing tools and provides AI-generated content suggestions based on approved brand guidelines."

AFTER (Audio-optimized): "Your brand has a voice. [SHORT PAUSE] Most content doesn't use it.

This platform changes that. It's the AI that knows your guidelines, your tone, your customer. [PAUSE]

Every piece of content it suggests? On-brand from the first word.

You don't edit for tone anymore. You edit for preference."


When to Apply This Skill

Content TypeApply?Notes
Blog audio versionYESFull 10 points
Training module narrationYESFull 10 points
Video script / voiceoverYESFull 10 points
Podcast show notesPARTIALPoints 4, 5, 6
Written blog post (no audio)NOUse normal writing conventions
Email copyNODifferent medium, different rules
Social media postsNOWritten medium

Quick Checklist

Before delivering any audio-destined script, verify:

  • Hook in first 10 words
  • At least 2 [PAUSE] markers
  • Average sentence length under 15 words
  • All contractions applied
  • No passive voice constructions
  • At least 2 emotional anchors
  • Reflection close at end
  • "You" appears at least 1x per paragraph

Built by PureBrain -- AI-powered marketing operations.

Want the full suite of 183+ production-tested skills? Start your trial ->

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.