Yt Description Writer

Community

sergebulaev

yt-description-writer

Write a full YouTube video description under 5,000 characters with a first-150-character hook (the only part visible before Show more, and what shows in search and suggested), timestamped chapters, naturally placed keywords, links, and a clear call to action. Covers long-form and YouTube Shorts. Becomes the create-post content that publishes with the video. Use to write or rework a description. Not for the title (use yt-title-optimizer) or the spoken opening (use yt-hook-scripter).

Overview

Publisher	sergebulaev
Repository	`youtube-skills`
Skill name	`yt-description-writer`
Stars	85
Forks	21
Bundled files	1
License	MIT
Links	Skill source Repository Homepage

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.
1 bundled files
Scripts, templates, and references the model can read while it works. Files are read-only and never executed.
Open source
Published by sergebulaev on GitHub. Read the source before you install it.

Installation

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

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.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/sergebulaev/youtube-skills/tree/main/skills/yt-description-writer
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.

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/sergebulaev/youtube-skills.git /tmp/youtube-skills
mkdir -p .claude/skills
cp -r /tmp/youtube-skills/skills/yt-description-writer .claude/skills/yt-description-writer

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Yt Description Writer 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 Yt Description Writer 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 Yt Description Writer 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.

YouTube Description Writer

The description does three jobs: it hooks the reader in its first 150 characters, it feeds YouTube and Google search, and it organizes the video with chapters and links. This skill writes all three layers.

When to use

User says "write the description for this video"
User has a video and wants chapters, keywords, and links structured
User wants the first 150 chars optimized for the search/suggested snippet
User is about to upload and needs the content that ships with the video

The three layers of a description

Layer	Chars	Job
Hook	first ~150	the only text visible before "Show more"; also the search/suggested snippet. Lead with the payoff or the question and the search phrase.
Body	~150 to 1,000	expand the value, restate the promise, place the main keywords naturally, link the key resource.
Furniture	rest, up to 5,000	chapters, links, social, credits, hashtags. Skimmable, labeled.

Steps

Voice profile first (all drafts). If ../../references/voice-profile.md has filled: yes, load it and match the user's voice fingerprint, hard rules, and CTA/link style throughout. If it is not filled, mention once that yt-community-post-writer --mode profile can learn their voice from a few posts, then proceed with the generic voice rules.

Never invent the specifics. The rules below ask for a concrete number, a date and a named entity, because that is what separates a real post from a generated one. Take them from what the user actually said in this conversation. Do not invent a figure, a date, a client name or a result, and do not soften a vague claim into a plausible-looking number. If the user has nothing concrete for a beat, ask them once, and if they still have nothing, drop the claim rather than decorate it. A published invented number is a retraction; a missing one is only a weaker post.

Gather inputs. The title (so the description complements it), the video's payoff, the main search phrase plus 3 to 5 secondary keywords, the chapter beats with timestamps, any links (lead magnet, gear, socials), and whether it is long-form or a Short. If a URL is given, parse it with lib/url_parser.py.
Write the first 150 chars as a second hook. Lead with the payoff or the question, work the primary keyword in naturally, and make it read well as a standalone snippet. This is not "welcome back to the channel".
Write the body. 2 to 4 short paragraphs that restate the promise and place the keywords without stuffing. Put the single most important link here.
Add chapters. Timestamps starting at 0:00, each label short and descriptive. YouTube needs at least 3 chapters and a 0:00 start to enable the chapter UI. For a Short, skip chapters.
Add the furniture. Links section, socials, a one-line CTA (subscribe for the specific reason, not "smash like"), and 3 to 5 relevant hashtags at the very end (the first 3 surface above the title).
Humanizer pass and length check. Scrub 2026 AI vocabulary by paragraph density (one marker is fine; three in one paragraph means rewrite it), cap em dashes at about one per 100 words (replace the excess with a comma, colon or line break, never a period), strip reveal bridges and sincerity openers ("let me be honest"); never add a hedge or confession the creator did not write. Keep the whole thing under 5,000 chars. Confirm the first 150 read well truncated.
Approval card. Show the description with a marked 150-char cutoff and the total char count.
On approval. This description is the content passed to lib.publish(kind="video"|"short", draft_text=<description>, title=<title>, ...). The title travels separately in platformSettings.youtube.title.

Hard rules

Global voice rules: see root SKILL.md Voice rules. Additional skill-specific rules:

5,000-char hard cap. The first 150 chars carry the hook and the search phrase.
The first line is never boilerplate ("Hi everyone, welcome back").
Chapters: at least 3, first one at 0:00, short labels. Long-form only.
Keywords appear in natural sentences, never as a comma-separated dump.
Hashtags: 3 to 5 max, at the end. More than 15 makes YouTube ignore all of them.
Every link is labeled so the reader knows what it is before clicking.

Anti-patterns (skill will refuse)

A boilerplate greeting in the first 150 chars.
A keyword dump or a tag wall pretending to be a description.
15+ hashtags (YouTube strips them all).
Em dashes above the cap (more than about one per 100 words), or an em dash swapped for a period.
"Smash that like button and subscribe" dead phrasing.
Chapters that do not start at 0:00 (the UI will not render them).
Raw unlabeled links.

Resources

../../references/algorithm-heuristics.md - the 150-char snippet, chapters, search signals
../../references/voice-rules.md - the canonical voice rules
references/description-template.md - the full skeleton with chapters, links, hashtags

Related skills

yt-title-optimizer - the title the description complements
yt-hook-scripter - the spoken first 30 seconds (different from the written hook)
yt-content-planner - keyword and link strategy across a series

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

references/description-template.md3 KB

Frequently asked questions

What does the Yt Description Writer AI skill do?

Why use Yt Description Writer on TypingMind?

Because you install it once and use it with any model. Yt Description Writer 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 Yt Description Writer in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/sergebulaev/youtube-skills/tree/main/skills/yt-description-writer. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Yt Description Writer?

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 Yt Description Writer?

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

Is the Yt Description Writer AI skill free?

Yes. It is published on GitHub by sergebulaev under the MIT license. 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

Brag

NousResearch

OrganizationPopular

Project launch video via Hyperframes, upstream-maintained.

Comfyui

Generate images, video, and audio via diffusion workflows.

Heartmula

HeartMuLa: Suno-like song generation from lyrics + tags.

Hyperframes

Render MP4/WebM videos from HTML compositions.

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

YouTube Description Writer

When to use

The three layers of a description

Steps

Hard rules

Anti-patterns (skill will refuse)

Resources

Related skills

Bundled files

Frequently asked questions

Related AI Skills

Set up your own AI workspace now