Skip to content
Back to search

youtube-skillbox

Free YouTube toolkit for AI agents — search, transcripts, community posts, home sections & Shows, plus 10 creator tools: scripts, hooks, SEO, chapters, audits. No API key, no credits, ever.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today
Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/hackers-reality--youtube-skillbox/badge.svg)](/listings/hackers-reality--youtube-skillbox)

HTML

<a href="/listings/hackers-reality--youtube-skillbox"><img src="/api/listings/hackers-reality--youtube-skillbox/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

**Requirements:** Python 3.11+ and [`yt-dlp`](https://github.com/yt-dlp/yt-dlp):

python -m pip install --user -U yt-dlp

> `posts`, `home`, `shows`, `search`, `keywords` and `track` work even *without* yt-dlp (pure HTTP). Everything else runs through it. Run it as a module (`python -m yt_dlp`) if the exe isn't on PATH.

Option 1: Install from PyPI (Recommended)

pip install youtube-skillbox

This installs the `ytdata` CLI and its bundled research engine. It does **not** install the 13 agent skill folders; use Option 3 to copy those into your agent's skills directory.

Option 2: Install from Source (Development)

git clone https://github.com/hackers-reality/youtube-skillbox.git
cd youtube-skillbox
pip install -e .

Option 3: Manual Skill Installation (for agents)

Point the installer at **your agent's skills directory** (where it discovers `SKILL.md` files):

./install.sh "/path/to/your/agents/skills"          # Linux / macOS — e.g. ~/.config/opencode/skills
.\install.ps1 "C:\path\to\your\agents\skills"       # Windows — e.g. %USERPROFILE%\.config\opencode\skills

…or by hand — copy any folder from [`skills/`](skills/) there yourself:

find skills -mindepth 1 -maxdepth 1 -type d ! -name yt-audit -exec cp -r {} ~/.config/opencode/skills/ \\;

Restart the agent session; the skills are auto-discovered by their `description` frontmatter.

---

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 420 lines
  1. <div align="center">
  2. <img src="assets/banner.svg" alt="YouTube Skillbox — 14 free YouTube skills for AI agents" width="100%">
  3. </div>
  4. <h1 align="center">YouTube Skillbox</h1>
  5. <p align="center">
  6. <b>14 free YouTube skills for AI agents — research, transcripts, and the full creator workflow.</b><br>
  7. No user-supplied YouTube Data API key · no credits · no expiry — <b>$0 forever</b>
  8. </p>
  9. <p align="center">
  10. <a href="#-table-of-contents">Table of Contents</a> ·
  11. <a href="#-demo">Demo</a> ·
  12. <a href="#-the-youtube-research-skill">Research</a> ·
  13. <a href="#-the-transcript-skill">Transcripts</a> ·
  14. <a href="#-when-yt-dlp-cant-do-it-we-build-it">Gap-fill engine</a> ·
  15. <a href="#%EF%B8%8F-the-10-creator-skills">Creator tools</a> ·
  16. <a href="#-install">Install</a> ·
  17. <a href="#-have-your-agent-set-it-up-for-you">Agent prompt</a>
  18. </p>
  19. <p align="center">
  20. <a href="https://pypi.org/project/youtube-skillbox/"><img src="https://img.shields.io/pypi/v/youtube-skillbox?label=PyPI&logo=pypi&logoColor=white" alt="PyPI version"></a>
  21. <a href="https://pypi.org/project/youtube-skillbox/"><img src="https://img.shields.io/pypi/pyversions/youtube-skillbox?logo=python&logoColor=white" alt="Python versions"></a>
  22. <a href="https://pypi.org/project/youtube-skillbox/"><img src="https://img.shields.io/pypi/l/youtube-skillbox?logo=opensourceinitiative&logoColor=white" alt="License"></a>
  23. <a href="https://github.com/hackers-reality/youtube-skillbox/actions/workflows/publish.yml"><img src="https://img.shields.io/github/actions/workflow/status/hackers-reality/youtube-skillbox/publish.yml?label=Publish&logo=githubactions&logoColor=white" alt="Publish status"></a>
  24. <a href="https://github.com/hackers-reality/youtube-skillbox/actions/workflows/tests.yml"><img src="https://img.shields.io/github/actions/workflow/status/hackers-reality/youtube-skillbox/tests.yml?label=Tests&logo=githubactions&logoColor=white" alt="Tests status"></a>
  25. <a href="https://github.com/hackers-reality/youtube-skillbox/releases"><img src="https://img.shields.io/github/v/release/hackers-reality/youtube-skillbox?display_name=tag&logo=github&logoColor=white" alt="GitHub release"></a>
  26. <a href="https://github.com/hackers-reality/youtube-skillbox/stargazers"><img src="https://img.shields.io/github/stars/hackers-reality/youtube-skillbox?style=social" alt="GitHub stars"></a>
  27. </p>
  28. ---
  29. ## ⚠️ Important Disclaimer
  30. **This tool accesses public YouTube data for research purposes only.**
  31. - Respect YouTube's [Terms of Service](https://www.youtube.com/t/terms) and [Community Guidelines](https://www.youtube.com/t/community_guidelines)
  32. - Respect rate limits — add delays between bulk requests
  33. - Not affiliated with YouTube or Google
  34. - For research/educational use only
  35. ---
  36. ## 📺 Demo
  37. Watch the agent in action — one prompt, *"why isn't this channel growing?"* OpenCode researches the videos, Shorts and transcripts, then surfaces the hooks, scripts and what to fix (why one Short hit 8K views while the rest couldn't pass 1K):
  38. <div align="center">
  39. <a href="https://www.youtube.com/watch?v=y9U_ZeJ713M" target="_blank">
  40. <img src="assets/thumbnail.png" alt="YouTube Skillbox Demo" width="80%">
  41. </a>
  42. <br>
  43. <sub>Click to watch on YouTube →</sub>
  44. </div>
  45. ---
  46. ## 📋 Table of Contents
  47. | Section | Description |
  48. |---------|-------------|
  49. | [📺 Demo](#-demo) | Video walkthrough of the agent using the skills |
  50. | [What is this?](#what-is-this) | Overview of the two-layer architecture |
  51. | [📺 The `youtube` research skill](#-the-youtube-research-skill) | 17 subcommands — search, keywords, compete, trends, track, scorecard, outliers, gap, channels, posts, shows |
  52. | [📜 The `transcript` skill](#-the-transcript-skill) | Free transcripts via yt-dlp |
  53. | [🛠 Gap-fill engine](#-when-yt-dlp-cant-do-it-we-build-it) | How we fill yt-dlp's gaps with raw scrapers |
  54. | [✂️ The 10 creator skills](#%EF%B8%8F-the-10-creator-skills) | Scripts, SEO, chapters, Shorts, and more |
  55. | [🧭 The `yt-guide` skill](#-yt-guide) | Skill router — when to use which skill, how to chain them |
  56. | [📦 Install](#-install) | Quick setup for any agent harness |
  57. | [🤖 Have your agent set it up for you](#-have-your-agent-set-it-up-for-you) | Copy-paste prompt for any coding agent |
  58. | [📝 Notes & limits](#-notes--limits) | Rate limits, pagination, edge cases |
  59. | [⚖️ Responsible use](#%EF%B8%8F-responsible-use) | Personal vs commercial use, risk notes |
  60. | [📄 License](#-license) | MIT License |
  61. | [🙏 Credits](#-credits) | yt-dlp, upstream skill sources |
  62. | [🧭 yt-guide skill](#-yt-guide) | Skill router — when to use which skill |
  63. ---
  64. ## What is this?
  65. A drop-in set of **[SKILL.md](https://github.com/anthropics/skills) skills** for agent harnesses like **OpenCode** (works anywhere skills are discovered from a folder). Each skill in [`skills/`](skills/) is self-contained: a `SKILL.md` the agent reads, plus small Python tools it runs.
  66. Two layers:
  67. | Layer | Skills | Powered by |
  68. | ----- | ------ | ---------- |
  69. | **Data** (research) | `youtube`, `transcript` | `yt-dlp` + custom raw YouTube scrapers |
  70. | **Creator workflow** | 10 × `yt-*` | local Python scripts, zero network deps for logic |
  71. | **Guide** | `yt-guide` | Skill router — when to use which skill |
  72. Everything is free. There is **no paid tier, no user-supplied YouTube Data API key, and no account** anywhere in the stack.
  73. ---
  74. ## 🔗 How the Skills Chain Together (The Pipeline)
  75. The skills aren't isolated — they form an **7-phase analysis pipeline** that agents can run end-to-end:
  76. ```mermaid
  77. graph TD
  78. A["User: Channel Analysis"] --> B["Phase 1: youtube"]
  79. B --> C["Phase 2: transcript"]
  80. C --> D["Phase 3: yt-script"]
  81. D --> E["Phase 4: yt-package"]
  82. E --> F["Phase 5: youtube (SEO/keywords)"]
  83. F --> G["Phase 6: youtube (outliers/gaps)"]
  84. G --> H["Phase 7: yt-retention*"]
  85. B --> J["Data: sections, videos, home, profile"]
  86. C --> K["Transcripts for top 3 videos"]
  87. D --> L["Hook scores 0-100"]
  88. E --> M["Title/thumbnail scores"]
  89. F --> N["Keywords, competition, scorecards"]
  90. G --> O["Outliers, competitor gaps"]
  91. H --> P["ONE fix + 3 working + don't-do"]
  92. ```
  93. *Phase 7 (yt-retention) optional — requires YouTube Studio CSV export.
  94. | Phase | Skill | Input | Output | Purpose |
  95. |-------|-------|-------|--------|---------|
  96. | 1 | `youtube` | @handle | Channel data (sections, videos, home, profile) | Data foundation |
  97. | 2 | `transcript` | Top 3 video IDs | Full transcripts (JSON + timestamps) | Raw content |
  98. | 3 | `yt-script` | Transcripts | Hook scores (0-100) per video | Retention hooks |
  99. | 4 | `yt-package` | Titles + hook scores | Title/thumbnail scores + briefs | CTR optimization |
  100. | 5 | `youtube` | Niche + target video | Keywords, competition, scorecards | Search visibility |
  101. | 6 | `youtube` | Channel + niche | Outliers, competitor gaps | Growth levers |
  102. | 7 | `yt-retention` | Studio CSV (optional) | Drop-off points, curves | Pacing fixes |
  103. **Each skill declares its pipeline position in its SKILL.md** — agents read this to know what comes before/after.
  104. **Final synthesis:** After running all phases, the agent synthesizes: **ONE biggest fix + 3 things working + what NOT to do.**
  105. ---
  106. ## 🧭 The `yt-guide` skill
  107. Not sure which skill to use? `yt-guide` is a **skill router** that tells the agent which skill to pick for any user request, how to use it, when to chain multiple skills, and common mistakes to avoid.
  108. **Decision tree (excerpt):**
  109. ```
  110. Channel audit / "why isn't this growing" → yt-workflow
  111. Plan content / calendar → yt-plan
  112. Write a script → yt-script
  113. Optimize a video for search → yt-seo (or youtube scorecard)
  114. Find Shorts in long-form video → yt-shorts
  115. Draft comment replies → yt-comment
  116. Edit decision list (cut dead air/filler) → yt-edit
  117. Analyze retention CSV → yt-retention
  118. ```
  119. **Chaining patterns:**
  120. | Pattern | Chain | Use when user asks… |
  121. |---------|-------|---------------------|
  122. | Channel Audit | `youtube → transcript → yt-script → yt-package → yt-seo → yt-viral → yt-retention` | "audit my channel" |
  123. | Video Creation | `yt-plan → yt-script → yt-package → yt-seo → [record] → yt-edit → yt-chapters → yt-shorts` | "help me create a video" |
  124. | SEO Optimization | `youtube info → yt-seo scorecard → yt-package title → yt-script hooks` | "optimize this video" |
  125. | Repurposing | `transcript → yt-shorts → yt-chapters → yt-package` | "turn this into Shorts" |
  126. | Research Only | `youtube search/keywords/compete → yt-viral` | "research this niche" |
  127. **For the agent:** Always read `yt-guide/SKILL.md` when unsure which skill to run. It maps user intent → skill and prevents the common mistake of running the full pipeline for a single-question request.
  128. ---
  129. ## 📺 The `youtube` research skill
  130. Seventeen subcommands behind one script — `skills/youtube/scripts/ytdata.py`:
  131. ```bash
  132. python skills/youtube/scripts/ytdata.py search "query" --limit 10 # find videos on a topic
  133. python skills/youtube/scripts/ytdata.py keywords "seed idea" --patterns # keyword ideas (2 engines), scored
  134. python skills/youtube/scripts/ytdata.py track "keyword" --video <url> # where your video ranks (+ history)
  135. python skills/youtube/scripts/ytdata.py profile "@handle" # About: subs, description, keywords
  136. python skills/youtube/scripts/ytdata.py sections "@handle" # count EVERY tab at once
  137. python skills/youtube/scripts/ytdata.py home "@handle" # home shelves + Featured Channels
  138. python skills/youtube/scripts/ytdata.py channel "@handle" --tab shorts
  139. python skills/youtube/scripts/ytdata.py latest "@handle" --limit 10
  140. python skills/youtube/scripts/ytdata.py posts "@handle" --limit 30 # community posts, paginated
  141. python skills/youtube/scripts/ytdata.py shows "@handle" # Shows tab (episode collections)
  142. python skills/youtube/scripts/ytdata.py playlist "<playlist-url>"
  143. python skills/youtube/scripts/ytdata.py info "<video-url-or-id>" # views, likes, tags, date
  144. ```
  145. `search`, `keywords`, `track` and `compete` take `--gl` / `--hl` (default `US` / `en`) to pin a
  146. region and language, so a rank position means the same thing on every machine.
  147. `keywords` merges two engines by default — YouTube's suggest plus DuckDuckGo's
  148. (`--sources yt` or `--sources ddg` to use one).
  149. Add `--json` to any command for structured output. Example:
  150. ```
  151. $ python skills/youtube/scripts/ytdata.py sections "@MrBeast"
  152. Channel: https://www.youtube.com/@MrBeast
  153. videos : 805
  154. shorts : 205
  155. streams : no tab
  156. shows : 5
  157. playlists: no tab
  158. community: 20+ posts (use `posts` to read them)
  159. ```
  160. | Command | Returns |
  161. | ------- | ------- |
  162. | `search` | Videos matching a query — title, duration, views, channel, URL (`--gl` / `--hl` locale) |
  163. | `keywords` | Keyword ideas from **YouTube suggest + DuckDuckGo** — a–z + pattern expansion, scored by cross-engine demand, `Y` = trending this year |
  164. | `compete` | Competition score (0–100) for a keyword — views, authority, title density, entrenchment, velocity |
  165. | `trends` | Historical trend for a seed keyword — score trajectory, velocity, sparklines (from `keywords` runs) |
  166. | `track` | Where a video ranks for a keyword — with ↑↓ history across runs, kept per locale |
  167. | `scorecard` | SEO checklist for a video — title/desc/tags/keyword fit, 0–100 score with actionable fixes |
  168. | `outliers` | Videos overperforming their channel — views/subscriber ratio, finds breakout content |
  169. | `gap` | Content gap analysis — my seed keywords vs competitor's profile keywords |
  170. | `profile` | Channel About section — name, handle, subs, description, **all** keywords |
  171. | `sections` | One-call counts of every tab: videos / shorts / streams / shows / playlists / posts |
  172. | `home` | Home-tab shelves — featured video, Shorts, playlists, posts, and **Featured Channels** with URLs |
  173. | `channel` | Channel uploads (`--tab videos \| shorts \| streams`) |
  174. | `latest` | Most recent uploads, reverse-chronological |
  175. | `posts` | Community posts — text, author, date, like count, URL. **Paginated** (~10/page) |
  176. | `shows` | Shows tab — title, episode count, URL |
  177. | `playlist` | Every video in a playlist |
  178. | `info` | One video's exact stats — views, likes, subs, date, tags, categories, description |
  179. ---
  180. ## 📜 The `transcript` skill
  181. Spoken text of any video, free via yt-dlp:
  182. ```bash
  183. python scripts/gettranscript.py "<url-or-id>" # clean text
  184. python scripts/gettranscript.py "<url-or-id>" --format stamped # [MM:SS] per line
  185. python scripts/gettranscript.py "<url-or-id>" --format json # segments + metadata
  186. python scripts/gettranscript.py "<url-or-id>" --lang es,es-419 # language priority list
  187. ```
  188. Prefers human captions, falls back to auto-generated. The JSON form carries `start`/`duration` per cue — ready for chapters, quotes, or edit lists.
  189. ---
  190. ## 🛠 When yt-dlp can't do it, we build it
  191. `yt-dlp` is the data engine for channels, playlists, metadata and transcripts — but it explicitly refuses a few things (returns `0 entries`, says `"no home tab"`, or quietly localizes results to your own IP). Rather than pay an API, each gap is filled with **a small pure-Python scraper** (`urllib` + JSON parsing of YouTube's own page data / Innertube endpoints):
  192. | Gap | What yt-dlp does | Our fill |
  193. | --- | ---------------- | -------- |
  194. | **Community posts** | `/community` → `0 entries` (wontfix) | Raw `ytInitialData` scrape of `/posts` (with `/community` fallback) |
  195. | **Post pagination** | — | **Innertube `browse` continuation POSTs** — the exact request infinite-scroll fires when you scroll. `--limit 30` reaches ~3 pages / months back |
  196. | **Home sections** | `"This channel does not have a home tab"` | Home-tab shelf parser: featured video, Shorts, playlists, posts, and **Featured Channels** with URLs |
  197. | **Shows tab** | `0 entries` (can't parse `gridShowRenderer`) | Direct renderer parser → title, episode count, show URL |
  198. | **Tab counts** | one tab per invocation | `sections` probes every tab and merges yt-dlp + raw results |
  199. | **Locale-pinned search** | search follows your IP — `gl` is silently set to whatever country you're in, and no flag exists to change it | Raw Innertube `search` POSTs with `client.gl` / `client.hl` pinned explicitly: same rank in New York and New Delhi. Works for `search`, `keywords`, `track`, `compete`, `trends`. |
  200. | **Keyword competition score** | not available | Live SERP analysis: median views (log), verified %, title match density, median age, **velocity (views/day)** → 0–100 score + band (EASY/MEDIUM/HARD/SATURATED) + actionable advice |
  201. | **Historical keyword trends** | not available | `keywords` auto-logs score snapshots to `~/.ytdata-trendlog.json`; `trends` shows sparklines, velocity (↑/↓/→), Δ over time |
  202. | **Video SEO scorecard** | not available | `scorecard` audits title/desc/tags/chapters/freshness vs target keyword → 0–100 with weighted checks & fixes |
  203. | **Channel outliers** | not available | `outliers` finds videos where views/subs ratio exceeds threshold — breakout content regardless of channel size |
  204. | **Content gap analysis** | not available | `gap` compares your seed keywords vs competitor's profile keywords → unique/shared opportunities |
  205. All of it is dependency-free Python — no API key, no credits, no tokens to manage.
  206. ```
  207. $ python scripts/ytdata.py home "@MKBHD" --limit 3
  208. 1. [featured video] Macbook PROLED XDR Impressions: ... (1 item)
  209. 2. [videos] Videos (12 items)
  210. 3. [featured channels] 🔥 MKBHD Approved 🔥 (6 items)
  211. - The Studio https://www.youtube.com/channel/UCG7J20...
  212. - Auto Focus https://www.youtube.com/channel/UC2J-0g...
  213. - WVFRM Podcast https://www.youtube.com/channel/UCEcrRX...
  214. ```
  215. ---
  216. ## ✂️ The 10 creator skills
  217. Local tools that turn raw transcripts/data into publishing artifacts:
  218. | Skill | Does |
  219. | ----- | ---- |
  220. | `yt-script` | Full video script from a raw idea — 21 hook formulas, scored |
  221. | `yt-package` | Title + thumbnail text written and linted as one pairing |
  222. | `yt-seo` | Description, tags and search-facing text aimed at real queries |
  223. | `yt-chapters` | Validated YouTube chapters straight from a transcript |
  224. | `yt-shorts` | Finds the self-contained Shorts hiding inside a long video |
  225. | `yt-edit` | Transcript → edit decision list (dead air, fillers, retakes) |
  226. | `yt-retention` | Reads a Studio retention export, says where viewers leave |
  227. | `yt-viral` | What's actually working in a niche, ranked by channel-beat |
  228. | `yt-plan` | A week/month content calendar sized to real capacity |
  229. | `yt-comment` | Drafts comment replies in the creator's voice, triaged |
  230. They pair naturally with the data layer: `youtube` finds the videos → `transcript` pulls the text → `yt-*` does the work.
  231. ---
  232. ## 📦 Install
  233. **Requirements:** Python 3.11+ and [`yt-dlp`](https://github.com/yt-dlp/yt-dlp):
  234. ```bash
  235. python -m pip install --user -U yt-dlp
  236. ```
  237. > `posts`, `home`, `shows`, `search`, `keywords` and `track` work even *without* yt-dlp (pure HTTP). Everything else runs through it. Run it as a module (`python -m yt_dlp`) if the exe isn't on PATH.
  238. ### Option 1: Install from PyPI (Recommended)
  239. ```bash
  240. pip install youtube-skillbox
  241. ```
  242. This installs the `ytdata` CLI and its bundled research engine. It does **not** install the 13 agent skill folders; use Option 3 to copy those into your agent's skills directory.
  243. ### Option 2: Install from Source (Development)
  244. ```bash
  245. git clone https://github.com/hackers-reality/youtube-skillbox.git
  246. cd youtube-skillbox
  247. pip install -e .
  248. ```
  249. ### Option 3: Manual Skill Installation (for agents)
  250. Point the installer at **your agent's skills directory** (where it discovers `SKILL.md` files):
  251. ```bash
  252. ./install.sh "/path/to/your/agents/skills" # Linux / macOS — e.g. ~/.config/opencode/skills
  253. ```
  254. ```powershell
  255. .\install.ps1 "C:\path\to\your\agents\skills" # Windows — e.g. %USERPROFILE%\.config\opencode\skills
  256. ```
  257. …or by hand — copy any folder from [`skills/`](skills/) there yourself:
  258. ```bash
  259. find skills -mindepth 1 -maxdepth 1 -type d ! -name yt-audit -exec cp -r {} ~/.config/opencode/skills/ \\;
  260. ```
  261. Restart the agent session; the skills are auto-discovered by their `description` frontmatter.
  262. ---
  263. ## 🤖 Have your agent set it up for you
  264. Don't want to install by hand? Paste this into any coding agent that can read files — **Claude Code, OpenCode, Cursor, or any agent with skill discovery**:
  265. ```text
  266. Set up my YouTube skills from https://github.com/hackers-reality/youtube-skillbox
  267. 1. Get the repo: git clone https://github.com/hackers-reality/youtube-skillbox.git
  268. (or fetch files from GitHub if you can't clone)
  269. 2. Install the data engine: python -m pip install --user -U yt-dlp
  270. (This is the ONLY runtime dependency — all skills share this one ytdata.py module)
  271. 3. Copy the data module to your agent's global skills directory (REQUIRED):
  272. - Linux/macOS: cp -r skills/youtube/scripts ~/.config/opencode/skills/youtube/
  273. - Linux/macOS: cp -r skills/transcript/scripts ~/.config/opencode/skills/transcript/
  274. - Windows: xcopy skills\youtube\scripts %USERPROFILE%\.config\opencode\skills\youtube\ /E /I
  275. - Windows: xcopy skills\transcript\scripts %USERPROFILE%\.config\opencode\skills\transcript\ /E /I
  276. (The SKILL.md files reference `scripts/ytdata.py` and `scripts/gettranscript.py`
  277. relative to their SKILL.md location. The scripts MUST be in the same folder structure.)
  278. 4. Point your agent at the skills:
  279. - **OpenCode / Cursor / local agents**: skills auto-discovered from
  280. .opencode/skills/ (project) and ~/.config/opencode/skills/ (global)
  281. - **Claude Code / web agents**: Add the skills/ folder path in settings,
  282. or copy skills/*/ to your agent's skills directory
  283. - **Any agent with "add skill" UI**: Point to skills/ folder or paste SKILL.md files
  284. 5. Smoke-test: run
  285. python ~/.config/opencode/skills/youtube/scripts/ytdata.py search "rick astley" --limit 1
  286. (should return 1 video with title, views, channel, URL)
  287. 6. Ask your agent: "List all 13 active YouTube skills and when to use each"
  288. It should read each skills/*/SKILL.md and answer from there.
  289. ```
  290. **Important**: The 13 active skills (youtube, transcript, 10 yt-* creator skills, and yt-workflow) share **one data module**: `skills/youtube/scripts/ytdata.py`. The deprecated `yt-audit` folder is intentionally skipped by both installers. The skills reference it via relative paths (`scripts/ytdata.py` from SKILL.md). You **must** copy the `scripts/` folder alongside each skill's SKILL.md in your agent's skills directory (step 3 above). An editable install exposes the `ytdata` CLI from your checkout; it does not copy skill folders into your agent's skills directory.
  291. ---
  292. ## 📝 Notes & limits
  293. - YouTube may throttle (HTTP 429) if you hammer it — a small delay between bulk calls is enough.
  294. - `posts` costs one HTTP request per ~10 posts; keep `--limit` ≤ 100 per invocation.
  295. - `sections` shows `N+` for community counts when more pages exist (exact when they fit on page 1).
  296. - Flat list view counts can show `?` occasionally — use `info` on a specific video for exact numbers.
  297. - Private / age-restricted / deleted targets fail with a clear message instead of silence.
  298. ## ⚖️ Responsible use
  299. This project reads publicly available YouTube data directly from the web rather than through the official API — strictly outside what YouTube's Terms of Service permit, even though that's common practice for personal tools like this.
  300. - **Personal use** — read-only public data, no login, low volume: low risk. Keep it that way.
  301. - **Commercial or high-volume use** — not recommended. Scaling it up, reselling data, or hammering YouTube meaningfully raises the risk of rate-limits, IP blocks, or takedown requests.
  302. - Never point it at private, age-restricted, or members-only content — the tools intentionally fail instead of bypassing.
  303. - **No user-supplied YouTube Data API key is required** — the project dynamically extracts YouTube's `INNERTUBE_API_KEY` from YouTube's public page configuration. This does not imply unlimited access, guaranteed uptime, unlimited requests, or immunity from YouTube rate limits.
  304. Use it at your own risk; this project is not affiliated with YouTube or Google. The code itself stays MIT-licensed — this section is guidance about how you use it against YouTube, not a restriction on the code.
  305. ## 📄 Third-Party Licenses
  306. This project includes third-party code under the following licenses:
  307. | Component | License | Copyright | Source |
  308. |-----------|---------|-----------|--------|
  309. | `youtube-agent-skill` (10 creator skills) | MIT | Copyright (c) 2026 Jake Schincariol | [Jakeschincariol/youtube-agent-skill](https://github.com/Jakeschincariol/youtube-agent-skill) |
  310. | `yt-dlp` | Unlicense | — | [yt-dlp/yt-dlp](https://github.com/yt-dlp/yt-dlp) |
  311. Full license texts are available in [`THIRD_PARTY_LICENSES/`](THIRD_PARTY_LICENSES/).
  312. ## 📄 License
  313. Released under the **[MIT License](LICENSE)** — free to use, modify, and redistribute.
  314. ## 🙏 Credits
  315. - **[yt-dlp](https://github.com/yt-dlp/yt-dlp)** — the extraction engine underneath the data layer.
  316. - The 10 creator (`yt-*`) skills come from the community GitHub collection **`youtube-agent-skill`** — installed here nearly as-is (deduplicated only), not rewritten from scratch.
  317. - The `youtube` and `transcript` skills (scripts and docs, including every gap-fill in this README) were built and rewritten in this project.
  318. - This project reads publicly available YouTube data; use it responsibly and respect YouTube's Terms of Service.
  319. <p align="center">Made so nobody has to pay for a YouTube API key again 🎉</p>

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.