New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Youtube

by unifapi-agent in unifapi-agent/agents

When a workflow needs public YouTube data through UnifAPI — channel, video, and playlist metadata, subscriber/view/like counts, video comments, caption tracks, channel community posts, related and…

MITAuto-check passedAgent Workflows

Install Youtube

skills CLI
$ npx skills add unifapi-agent/agents --skill youtube -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install unifapi-agent/agents youtube --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/unifapi-agent/agents.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/youtube-agent/youtube .claude/skills/youtube && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
youtube
GitHub stars
587
Token cost
~2.6k tokens
SKILL.md length
887 words
Files
2
Skills in repo
47
Repo updated
First seen
Licence
MIT

At a glance

When a workflow needs public YouTube data through UnifAPI — channel, video, and playlist metadata, subscriber/view/like counts, video comments, caption tracks, channel community posts, related and…

  • Works in 8 steps: Frame the query. Call… → Resolve a channel. Call… → Read a channel's output. Call… → …
  • Tasks that involve Skill authoring
  • SKILL.md covers Use the unifapi skill for live…, Response contract, Core operations and Workflow, plus 4 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Youtube is an agent skill from unifapi-agent/agents. When a workflow needs public YouTube data through UnifAPI — channel, video, and playlist metadata, subscriber/view/like counts, video comments, caption tracks, channel community posts, related and trending videos, hashtag videos, and search across videos, channels, and playlists. Also use on "who owns this topic on YouTube," "pull this channel's videos," "YouTube view counts," "what are commenters saying," "get this video's captions/transcript," "this channel's playlists," or when another skill (creator vetting…

Its SKILL.md is about 2.6k tokens, which your agent loads only when the skill is triggered. The skill folder holds 1 other file (for example `README.md`).

It sits in Agent Workflows, covering Skill authoring. It works with YouTube. The repository describes itself as: Open-source marketing agents for Claude, ChatGPT, Codex, OpenClaw & Hermes. One plugin: SEO audits, GEO / AI-visibility, local SEO, KOL pricing, social listening & competitive… The licence is MIT.

When your agent uses it

  • Tasks that involve Skill authoring

Example prompts

  • “who owns this topic on YouTube,”
  • “pull this channel”
  • “YouTube view counts,”
  • “/youtube”

Workflow steps

8 steps, taken from the first numbered list in SKILL.md.

  1. Frame the query. Call youtube/search?q=... for who owns a topic, or
  2. Resolve a channel. Call youtube/resolve/channel-id?url=... to get the
  3. Read a channel's output. Call youtube/channels/{channel_id}/videos (and
  4. Read a video and its neighborhood. Call youtube/videos/{video_id} for
  5. Read audience and transcript. Call youtube/videos/{video_id}/comments
  6. Walk a playlist or hashtag. Call youtube/playlists/{playlist_id} then
  7. Size topic demand. Aggregate view_count across the search and related
  8. Cite everything. Every figure ties back to the video or channel it came

What it can do on your machine

Read from SKILL.md and the folder at commit fb53247. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are json and markdown).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Youtube loads about 2.6k tokens when it runs. Until then it costs about 170 tokens; SKILL.md has 887 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~170
When it runs · the whole SKILL.md, loaded when a task matches
~2.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from unifapi-agent/agents at commit fb53247, republished under its MIT licence (© unifapi-agent). 887 words, ~2,610 tokens.

Download SKILL.mdSave it as .claude/skills/youtube/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
youtube
description
When a workflow needs public YouTube data through UnifAPI — channel, video, and playlist metadata, subscriber/view/like counts, video comments, caption tracks, channel community posts, related and trending videos, hashtag videos, and search across videos, channels, and playlists. Also use on "who owns this topic on YouTube," "pull this channel's videos," "YouTube view counts," "what are commenters saying," "get this video's captions/transcript," "this channel's playlists," or when another skill (creator vetting, competitor reception, content demand) needs the deterministic YouTube read path. Connect via the `unifapi` skill first. Read-only research, never uploads.
license
MIT
metadata.author
UnifAPI
metadata.version
1.0.0
metadata.homepage
https://unifapi.com/agents/youtube
metadata.source
https://github.com/unifapi-agent/agents

youtube

The deterministic read path for public YouTube data through UnifAPI. This is a Data Skill: it does not run a marketing job on its own — it names the concrete youtube/... operations, response shapes, and the real coverage limits so any channel- or content-first workflow (creator vetting, competitor reception, content demand) reads from one known recipe instead of rediscovering the surface each time.

Read-only — eyes, not hands. It researches public YouTube data and returns cited records; it never uploads, comments, or subscribes, and UnifAPI never holds YouTube credentials.

Use the unifapi skill for live evidence

Connect once through the shared unifapi skill (OAuth MCP), then call the operations below. Channels are keyed by {channel_id} and videos by {video_id} — resolve a handle or URL to a channel id first (see Gotchas). Keep any billing metadata so the output can state record cost.

Response contract

Single-entity endpoints return the object in data:

json
{
  "request_id": "unif_...",
  "data": {},
  "billing": { "records_charged": 1, "balance_remaining": 99 }
}

List endpoints return an array in data plus pagination:

json
{
  "request_id": "unif_...",
  "data": [],
  "pagination": { "has_more": false, "next_cursor": null },
  "billing": { "records_charged": 1 }
}

When pagination.has_more is true, pass pagination.next_cursor as the next request's cursor. Always preserve billing when reporting cost.

Core operations

NeedOperation
Search videosyoutube/search (?q=..., optional sort_by/upload_date/duration/region/language)
Search channels / playlistsyoutube/search/channels · youtube/search/playlists (?q=...)
Trendingyoutube/trending (optional type = now/music/gaming/movies)
Hashtag videosyoutube/hashtags/{tag}/videos
Resolve handle/URL → idyoutube/resolve/channel-id (?url=...)
Channel pageyoutube/channels/{channel_id}
Channel videos / shortsyoutube/channels/{channel_id}/videos · .../shorts
Channel playlistsyoutube/channels/{channel_id}/playlists
Channel community postsyoutube/channels/{channel_id}/community
Search within a channelyoutube/channels/{channel_id}/search (?q=...)
Video metadatayoutube/videos/{video_id}
Related videosyoutube/videos/{video_id}/related
Video commentsyoutube/videos/{video_id}/comments (optional sort_by = top/newest)
Video captions / transcriptyoutube/videos/{video_id}/captions
Playlist page / videosyoutube/playlists/{playlist_id} · .../videos

Need a field not listed here? Use the unifapi skill's get_operation to read the exact schema before calling — but pick the operation from this table, don't discover blind.

Workflow

The deterministic recipes. Pick the one that matches the job; each names exactly what to call.

  1. Frame the query. Call youtube/search?q=... for who owns a topic, or youtube/trending for what's surging now.
  2. Resolve a channel. Call youtube/resolve/channel-id?url=... to get the channel_id, then youtube/channels/{channel_id} for subscriber_count, video_count, and view_count.
  3. Read a channel's output. Call youtube/channels/{channel_id}/videos (and .../shorts); each video's view_count is the demand signal. Page via next_cursor. youtube/channels/{channel_id}/playlists maps how the catalog is organized; youtube/channels/{channel_id}/community reads announcements and engagement outside videos; youtube/channels/{channel_id}/search?q=... finds a creator's coverage of one topic without scanning the whole catalog.
  4. Read a video and its neighborhood. Call youtube/videos/{video_id} for metadata (now including like_count, category, published_at, has_captions), then youtube/videos/{video_id}/related to map adjacent demand and competing content.
  5. Read audience and transcript. Call youtube/videos/{video_id}/comments (sort_by top/newest) for what the audience says — raw comment text, like and reply counts; any sentiment or FAQ grouping is yours to compute, the API returns text only. Call youtube/videos/{video_id}/captions for caption tracks (timed-text url per language) to read or analyze the transcript.
  6. Walk a playlist or hashtag. Call youtube/playlists/{playlist_id} then .../videos to ingest a curated series in order; youtube/hashtags/{tag}/videos tracks a campaign or trend across creators.
  7. Size topic demand. Aggregate view_count across the search and related results for a topic; title, description, and keywords show how the topic is framed.
  8. Cite everything. Every figure ties back to the video or channel it came from; report billing.records_charged (or estimate when billing metadata is absent).
Show full SKILL.md (365 more words)Show less

Shape notes

  • YouTubeChannel — keyed by {channel_id}. subscriber_count, video_count, view_count, handle, keywords, links, country, created_at. is_verified is not populated by the upstream and is always false — don't report verification you can't read.
  • YouTubeVideo — keyed by {video_id}. view_count, like_count (best-effort; 0 when the creator hides it), category, channel_id, author, duration_seconds, keywords, published_at, has_captions, is_live, is_private.
  • YouTubeVideoPreview — search/related/list rows: adds published_time. category is part of the shape but empty on list rows (only YouTubeVideo detail carries it).
  • YouTubeComment — text, author, author_channel_id, like_count, reply_count, published_time, is_verified, is_creator. Text only — no sentiment field.
  • YouTubeCaptions — video_id, tracks[] (language_code, timed-text url, is_translatable), translation_languages[]. Single object, billed as one record. Track urls are time-limited.
  • YouTubePlaylist / YouTubePlaylistPreview — playlist title, channel_id, author, video_count (detail adds view_count, last_updated; preview adds first_video_id).
  • YouTubeChannelPreview — search-channels rows: title, handle, description, subscriber_count, video_count.
  • YouTubeCommunityPost — text, author, vote_count, reply_count, attachment_type, image_url.

Gotchas

  • Resolve only goes to channel ids. youtube/resolve/channel-id?url=... resolves a handle or URL to a {channel_id} — call it before any youtube/channels/{channel_id}/... call. There is no video- or playlist-id resolver; pass a known {video_id} / {playlist_id} (parse it from the URL yourself).
  • Comments are raw text. youtube/videos/{video_id}/comments returns comment text plus like/reply counts — no sentiment, no FAQ grouping. Compute those yourself from the text; don't claim a sentiment field the API doesn't return.
  • like_count is best-effort. YouTubeVideo.like_count is 0 when the creator hides likes — treat 0 as "hidden or none," not authoritative.
  • is_verified is always false for channels (upstream doesn't expose it) — don't report channel verification.
  • Captions can be empty. A video with captions disabled returns tracks: []; check before promising a transcript. Track urls are time-limited.
  • Hashtags take the tag with or without a leading # (youtube/hashtags/lofi/videos).
  • A low balance can silently truncate list pages: check billing.truncated_due_to_balance — when true the page is partial.

Output

Return the records the calling workflow needs, each cited to its video or channel, plus a one-line cost note (records_charged). When this skill is used directly, a compact demand brief is the default:

markdown
**{Channel}** — {subscribers} subscribers, {videos} videos, {totalViews} views. Top videos for {topic}: {title} ({views} views). Related/competitors: {channels}. Evidence: {URLs}. Records: ~{N}.
  • kol-pricing, creator-shortlist, audience-fit-check (Influencer Marketing) — price and vet YouTube channels on top of this read path.
  • competitor-profiling (Competitive Intelligence), content-opportunity-brief (Content Strategy) — reception and content-demand work that reads YouTube first.
  • unifapi — the shared data skill: connect MCP and look up exact schemas with get_operation.

© unifapi-agent, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in skills/youtube-agent/youtube of unifapi-agent/agents.

  • SKILL.md
  • README.md

Open the folder on GitHubat commit fb53247

Compare with similar skills

Youtube next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Youtube compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Youtube this skillunifapi-agent/agents587—~2.6kAutomated safety check: PassMIT
Yt Briefjeremylongshore/tons-of-skills-marketplace2.8k—~1.5kAutomated safety check: PassMIT
Youtube Strategyericrisco/rsc-harness167—~3.1kAutomated safety check: PassMIT
Skill CreatorAzure/azqr79589 repos~8.2kAutomated safety check: PassApache-2.0
Claude Code Skill Developer Guidediet103/claude-code-infrastructure-showcase10k11 repos~3.5kAutomated safety check: PassMIT
Banner Design Systemnextlevelbuilder/ui-ux-pro-max-skill134k1 repos~1.8kAutomated safety check: PassMIT

Similar skills

  • Yt Brief

    jeremylongshore/tons-of-skills-marketplace

    Refine a YouTube video idea into a structured production brief with angle, key points, value proposition, CTA asset, and audience segment.

    2.8k GitHub stars~1.5k tokensUpdated today
    Agent WorkflowsAuto-check passed

  • Youtube Strategy

    ericrisco/rsc-harness

    A skill your agent uses when a YouTube channel needs a system-level decision — positioning and niche, upload cadence, how videos chain into playlists and series, end-screen routing, or a…

    167 GitHub stars~3.1k tokensUpdated yesterday
    Media & CreativeAuto-check passed

  • Skill Creator

    Azure/azqr

    Official

    Create new skills, modify and improve existing skills, and measure skill performance.

    795 GitHub starsUsed in 89 repos~8.2k tokens
    Agent WorkflowsAuto-check passed

  • Claude Code Skill Developer Guide

    diet103/claude-code-infrastructure-showcase

    A guide to creating and managing Claude Code skills with auto-activation: skill-rules.json triggers, hooks, enforcement levels, YAML frontmatter and progressive disclosure.

    10k GitHub starsUsed in 11 repos~3.5k tokens
    Agent WorkflowsAuto-check passed

  • Banner Design System

    nextlevelbuilder/ui-ux-pro-max-skill

    Walks through designing a banner for social media, ads, a website hero or print, from gathering requirements to building 2 or 3 art-direction options in HTML and CSS.

    134k GitHub starsUsed in 1 repo~1.8k tokens
    Frontend & DesignAuto-check passed

  • Nuwa Thinking-Style Skill Builder

    alchaincyf/nuwa-skill

    Researches a person or theme and distills how they think into a runnable persona skill with mental models, decision rules and a characteristic voice.

    34k GitHub stars~4.6k tokensUpdated 1 mo ago
    Agent WorkflowsAuto-check passed

More from unifapi-agent/agents

All 47 skills in this repo

  • LLM Mention Tracking

    unifapi-agent/agents

    When the user wants to track how often their brand or domain gets mentioned across ChatGPT and AI search engines over a set of prompts, and how that share of voice compares to named competitors over…

    587 GitHub stars~1.8k tokensUpdated 1 mo ago
    Auto-check passed

  • Buying Signal Monitor

    unifapi-agent/agents

    When a seller or SDR wants to catch public buying intent on X/Twitter and LinkedIn — someone asking for a tool they sell, complaining about or switching off a competitor, or hiring for a role that…

    587 GitHub stars~2.4k tokensUpdated 1 mo ago
    Auto-check passed

  • Customer Research

    unifapi-agent/agents

    When the user wants to research customers from public communities, or synthesize customer language, pains, and objections.

    587 GitHub stars~2.1k tokensUpdated 1 mo ago
    Auto-check passed

  • Keyword Research

    unifapi-agent/agents

    When the user wants to research keywords, find keyword opportunities, run a keyword gap analysis, build topic clusters, or compare what competitors rank for.

    587 GitHub stars~2.7k tokensUpdated 1 mo ago
    Auto-check passed

  • Schema

    unifapi-agent/agents

    When the user wants to add, fix, or optimize schema markup and structured data on their site.

    587 GitHub stars~2.1k tokensUpdated 1 mo ago
    Auto-check passed

  • SEO Audit

    unifapi-agent/agents

    When the user wants to audit, review, or diagnose SEO issues on their site.

    587 GitHub stars~3.4k tokensUpdated 1 mo ago
    Auto-check passed

Works with

Categories

Questions about Youtube

What does Youtube do?

When a workflow needs public YouTube data through UnifAPI — channel, video, and playlist metadata, subscriber/view/like counts, video comments, caption tracks, channel community posts, related and…. Youtube is an agent skill from unifapi-agent/agents. When a workflow needs public YouTube data through UnifAPI — channel, video, and playlist metadata, subscriber/view/like counts, video comments, caption tracks, channel community posts, related and trending videos, hashtag videos, and search across videos, channels, and playlists.

When should I use Youtube?

Youtube fits situations like: tasks that involve Skill authoring.

How do I install Youtube in Claude Code?

Run `npx skills add unifapi-agent/agents --skill youtube -a claude-code`. Or copy the skill folder (skills/youtube-agent/youtube in unifapi-agent/agents) into .claude/skills/youtube in your project. Claude Code loads it when a task matches its description.

How do I install Youtube in Codex?

Run `npx skills add unifapi-agent/agents --skill youtube -a codex`. Or copy the skill folder (skills/youtube-agent/youtube in unifapi-agent/agents) into .agents/skills/youtube in your project. Codex loads it when a task matches its description.

Can I use Youtube in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add unifapi-agent/agents --skill youtube -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/youtube, .gemini/skills/youtube, .github/skills/youtube and .opencode/skills/youtube in your project.

What does Youtube need to run?

SKILL.md names no scripts, command-line tools or credentials: Youtube is instructions for the agent only.

Does Youtube access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Youtube safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Youtube use?

Youtube is published under the MIT licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Youtube use?

About 2.6k tokens (SKILL.md is roughly 10k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Youtube?

Skills that share tags, products or a category with Youtube: Yt Brief (jeremylongshore/tons-of-skills-marketplace, 2.8k stars), Youtube Strategy (ericrisco/rsc-harness, 167 stars), Skill Creator (Azure/azqr, 795 stars) and Claude Code Skill Developer Guide (diet103/claude-code-infrastructure-showcase, 10k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Youtube?

unifapi-agent (a GitHub organization) maintains it in unifapi-agent/agents, which has 587 GitHub stars. The repository holds 47 skills in this directory. The repository was last updated on September 5, 2026.

Source: unifapi-agent/agents on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.