Kai Video logo

Kai Video

Community
cgallic
kai-video

Produce video scripts and clipping plans for TikTok, YouTube Shorts, Instagram Reels, and long-form YouTube. Generates hook-first scripts optimized for each platform's algorithm, plus a clipping plan to extract short-form from long-form content. Use when "video script", "TikTok script", "YouTube script", "Reels script", "Shorts script", "video content", "clipping plan", "video ideas", or any request to create video content for social platforms.

Overview

Publishercgallic
Repositorykai-cmo-harness
Skill namekai-video
Stars
61
Forks
8
Bundled files
Instructions only
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by cgallic on GitHub. Read the source before you install it.

Installation

Install the Kai Video AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/cgallic/kai-cmo-harness/tree/main/harness/skills-v2/kai-video
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/cgallic/kai-cmo-harness.git /tmp/kai-cmo-harness
mkdir -p .claude/skills
cp -r /tmp/kai-cmo-harness/harness/skills-v2/kai-video .claude/skills/kai-video
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Kai Video in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Kai Video on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Kai Video is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

/kai-video — shootable scripts and a defensible clip list

Kai root note: knowledge/, harness/, and scripts/ paths in this skill live in the Kai install, not the user's project. Resolve them against the first ancestor directory of this SKILL.md that contains a knowledge/ folder (the Kai plugin root, ~/.claude/kai, or the kai-cmo-harness repo). MARKETING.md, memory/, and any output files live in the current project. If a referenced scripts/ command is not available in this install, say so, skip it, and continue with the file-based guidance — never fabricate its output.

Objective

Scripts someone can shoot without asking a follow-up question, and — where long-form source exists — a clipping plan naming which moments become short-form, at which timestamps, and why. Hook-first and platform-specific: the first three seconds decide whether the rest of the work matters.

Done when

Work type social-post — floor E5/C2/O3 (harness/eco-floors.yaml), contract harness/skill-contracts/social-post.yaml.

  • E5 — the video is live and a non-actor read the public permalink back against the approved script and cut. A script file in workspace/ is E1; approval of the exact script is E3. This skill reaches the artifact; publishing is a separate, approved step.
  • C2 — four_us_score at 10/16 and banned_word_check clean on the script copy, plus the per-script bar below.
  • O3 — reach, engagement rate, profile clicks, and link clicks read from platform insights at 7 days, against a baseline recorded before publish.

Per-script bar: hook lands in the first 3 seconds (no intros, no "hey guys", no logo animation) · one idea per short-form video · zero banned words and zero AI slop · a CTA that closes naturally rather than being bolted on · visual direction included, not words alone · length appropriate to the platform.

Constraints

  • Read MARKETING.md from the project root before asking the user anything. If it does not exist, build it from the codebase — CLAUDE.md, README.md, PROJECT.md, package.json, landing pages, email/ad/analytics config — using the template from /kai-email-system, and confirm the draft.
  • Know these before scripting: topic, target platforms, format (talking head, screen recording, b-roll, animation), length band (15–60s short-form or 5–15min long-form), goal, and any existing content to adapt from.
  • Load the platform guide before writing for that platform — the algorithms differ and the pacing rules are not transferable.
  • Every clip row carries a source timestamp or file locator. If the source is a guest, customer, podcast, or third-party show, mark approval status before export and follow harness/references/transcript-video-research-rules.md.
  • Clip plans are dry runs. Publishing or upload requires human approval.
  • Vary the hook formula across videos. Repeating one formula across a slate trains the audience to skip.

Context

NeedLoad
Video content mechanicsknowledge/playbooks/video-content-creation.md
TikTokknowledge/channels/tiktok-algorithm.md
YouTube (Shorts and long-form)knowledge/channels/youtube.md
Instagram Reelsknowledge/channels/instagram.md
Clipping long-form into short-formknowledge/playbooks/video-clipping-automation-workflow.md
Any transcript, caption, podcast, webinar, or third-party sourceharness/references/transcript-video-research-rules.md
Product, ICP, voice, channelsMARKETING.md (project root)

Short-form script (15–60s) carries: platform, length, format, HOOK (0–3s, pattern interrupt — question, bold claim, or visual surprise), BODY (3–45s, one idea and one takeaway with bracketed visual direction), CTA (last 3–5s), timestamped on-screen text overlays, 3–5 platform-appropriate hashtags, and a trending-audio suggestion or "original audio".

Long-form script (5–15min) carries: HOOK (0–30s previewing the payoff), INTRO (30s–1min: context, credibility, structure promise), three content sections with visual direction, RECAP + CTA, chapter timestamps, and a thumbnail concept (3–5 words of title text, expression, key visual).

Hook formulas — rotate across a slate:

TypeFormulaExample
Question"Did you know [surprising fact]?""Did you know 73% of leads call after hours?"
Bold claim"[Contrarian statement]""Cold email is dead. Here's what replaced it."
Problem"If you're struggling with [X]...""If your landing page converts under 3%..."
Result"We went from [bad] to [good] in [time]""We went from 2% to 11% conversion in 3 weeks"
List"[Number] [things] that [outcome]""5 emails every SaaS needs on day one"

Clip scoring — score every candidate before scripting it, 1–5 per factor:

FactorQuestion
HookDo the first 3 seconds create a clear reason to keep watching?
Standalone clarityCan the viewer understand the moment without the full source?
ProofDoes it include a concrete example, result, quote, or demo moment?
TensionDoes it contain contrast, surprise, objection, stakes, or a reveal?
Source safetyIs the claim attributable and cleared for use?

Clips scoring 18/25 or higher are production candidates. Lower scores go to workspace/video/clipping-plan/_kill-list.md with the reason.

Output goes to workspace/video/: _video-plan.md, short-form/ (per-platform scripts), long-form/, clipping-plan/clips-from-[source].md, _quality-report.md.

Escalate when

  • Source footage rights, guest approval, or customer permission for a clip is unconfirmed.
  • A claim in the script needs a number that has no source.
  • The requested length or format conflicts with the platform's current norms in the loaded channel guide.
  • The user wants the video rendered rather than scripted — that is /kai-video-production.
  • Anything would be uploaded or published before human approval.

Frequently asked questions

What does the Kai Video AI skill do?

Produce video scripts and clipping plans for TikTok, YouTube Shorts, Instagram Reels, and long-form YouTube. Generates hook-first scripts optimized for each platform's algorithm, plus a clipping plan to extract short-form from long-form content. Use when "video script", "TikTok script", "YouTube script", "Reels script", "Shorts script", "video content", "clipping plan", "video ideas", or any request to create video content for social platforms.

Why use Kai Video on TypingMind?

Because you install it once and use it with any model. Kai Video is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Kai Video in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/cgallic/kai-cmo-harness/tree/main/harness/skills-v2/kai-video. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Kai Video?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Kai Video?

As many as you like. As long as a model supports skills, you can use Kai Video with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Kai Video AI skill free?

It is published on GitHub by cgallic. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇