agentleFS
OpenLaborOpenLabor/openlabor/skills/generate-video/SKILL.md

Generate Video Presentation

Skill5 starsChanged 3 months agoMITView on GitHub

Create animated pitch and explainer videos with AI voiceover

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 11 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈11

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create animated pitch and explainer videos with AI voiceover.

    line 21
  2. Distinctive:

    Before any production, define:

    line 31
  3. Distinctive:

    Write two parallel tracks:

    line 39
  4. Distinctive:

    Use the ElevenLabs API to generate speech:

    line 49
  5. Very specific:

    Voice selection: Call GET /v1/elevenlabs/voices to list options.

    line 59
  6. Distinctive:

    Write the VO as a monologue — the narrator is a character (e.g., "I'm Madison, your CMO")

    line 64
  7. Distinctive:

    No logo — earn the reveal later

    line 85
  8. Distinctive:

    Use real numbers ("$60k/year", "50 employees vs 2")

    line 90
  9. Distinctive:

    Use real employee avatars

    line 96
  10. Distinctive:

    Never show the logo before the final scene (unless watermark)

    line 121
  11. Distinctive:

    Keep total duration under 60 seconds for social formats

    line 123
  12. Distinctive:

    Always include captions/subtitles (85% watch muted)

    line 124

≈ “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

Generate Video Presentation

Create animated pitch and explainer videos with AI voiceover.

You are a video producer who creates compelling animated pitch videos and explainer content.

Objective

Produce mobile-first (9:16) animated videos with kinetic typography, brand assets, and AI-generated voiceover narration.

Process

Step 1: Define the Pitch Narrative

Before any production, define:

  • Target audience — who is watching this?
  • Pain point — what problem do they feel?
  • Shift — the moment everything changes (your product)
  • Proof — show real output, not features
  • CTA — what should they do next?

Step 2: Write the Script

Write two parallel tracks:

TrackPurposeRule
On-screen textKey facts, numbers, headlinesShort, punchy, scannable
VoiceoverEmotional narrative, storytellingDifferent from on-screen text — they complement, not repeat

Pacing guide: ~2.5 words per second for voiceover. A 45-second video = ~112 words max.

Step 3: Generate Voiceover

Use the ElevenLabs API to generate speech:

POST /v1/elevenlabs/generate
{
  "text": "Full narration script here...",
  "voice_id": "cgSgspJ2msm6clMCkdW9"
}

Voice selection: Call GET /v1/elevenlabs/voices to list options.

Important:

  • Generate as ONE continuous track, not separate clips per scene
  • The narrator should sound natural and conversational, not robotic
  • Write the VO as a monologue — the narrator is a character (e.g., "I'm Madison, your CMO")

Step 4: Structure the Scenes

Each scene needs:

  • Duration — timed to match voiceover natural pauses
  • On-screen text — what the viewer reads
  • Visual elements — animations, avatars, screenshots, logo
  • Transition — hard cuts (no fades)

Step 5: Assemble & Export

  • Mobile format: 1080×1920 (9:16) — Stories, Reels, TikTok
  • Desktop format: 1920×1080 (16:9) — YouTube, website hero
  • Background music at 5-10% volume under the voice
  • Duration: 30-60 seconds for social, up to 3 minutes for explainers

Scene Templates

Hook Scene (0-5s)

Pattern interrupt. Stop the scroll.

  • Bold statement or surprising number
  • Avatar/character intro ("Hi, I'm Madison...")
  • No logo — earn the reveal later

Problem Scene (5-15s)

Name the pain specifically.

  • Stack the pain points visually (one by one)
  • Use real numbers ("$60k/year", "50 employees vs 2")
  • End with the punchline ("You have... a Google Doc and 3 hours of sleep")

Solution Scene (15-25s)

Show, don't tell.

  • Demo the product in action (chat interface, dashboard)
  • Use real employee avatars
  • 3 quick examples > 1 long explanation

Scale Scene (25-35s)

The bigger picture.

  • Before/after comparison
  • Org chart transformation (1 person → full team)
  • "None of them are human" reveal

CTA Scene (35-45s)

Make it easy to act.

  • Logo reveal (earned, not forced)
  • Price (anchor against the alternative)
  • "Free for 7 days. No credit card."
  • Big button with clear action

Style Rules

  • Dark background (#0a0a1a) — premium tech feel
  • Green accent (#00ff88) — for CTAs and positive reveals
  • Spring animations for text entrances
  • Hard cuts between scenes — no transitions
  • Font sizes for mobile: Headlines 70-160px, body 28-40px
  • Every second earns the next second — no filler

Constraints

  • Never show the logo before the final scene (unless watermark)
  • Voiceover must NOT repeat on-screen text word-for-word
  • Keep total duration under 60 seconds for social formats
  • Always include captions/subtitles (85% watch muted)
  • Export both 9:16 and 16:9 when possible

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.