Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/HasData/youtube-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/rules/hasdata/youtube-mcp/hasdata-youtube)<a href="https://agentmods.dev/rules/hasdata/youtube-mcp/hasdata-youtube"><img src="https://agentmods.dev/badge/rules/hasdata/youtube-mcp/hasdata-youtube/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/hasdata/youtube-mcp/hasdata-youtube"><img src="https://agentmods.dev/badge/rules/hasdata/youtube-mcp/hasdata-youtube.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00019 | $0.01005 |
| Opus 5.5 | $0.00008 | $0.00402 |
| Sonnet 5.5 | $0.00004 | $0.00201 |
| Haiku 4.5 | $0.00002 | $0.00101 |
Grade A, and why
hasdata-youtube scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 49 lines — stays where its author put it; the contents beside it link to each section on GitHub.
YouTube through HasData
Four tools. hasdata_youtube_search_getYoutubeSearchResults takes q, hasdata_youtube_video_getYoutubeVideo and hasdata_youtube_transcript_getYoutubeTranscript take a video id as v, and hasdata_youtube_channel_getYoutubeChannel takes channelId, which also accepts an @handle.
The search page is five lists, not one
Results split into videoResults, inlineShortsResults, playlistResults, adsResults and sponsoredResults. Reading only the first misses the shorts and the playlists, and folding the last two into the organic set reports paid placements as rankings.
searchInformation.totalResults sizes the query, and pagination.nextPageToken goes back in as paginationToken for the next page.
sortBy takes relevance, popularity, rating, date or views. date narrows to hour, today, week, month or year, videoType to video, shorts, channel, playlist or movie, and length to under4, between420 or plus20. filters__, spelled with two underscores, takes flags including subtitles, which is how to keep only videos that have a transcript to read.
Counts use two opposite conventions
In the search and channel payloads views is the number and viewsOriginal is the printed string. The video call reverses it, where views reads "482,284 views" and extractedViews holds 482284, the same way as extractedLikes and channel.extractedSubscribers. Read the type rather than the name, and sort on whichever one is numeric.
publishedDate is a relative string such as 4 years ago, so it cannot be turned into a date. A window belongs in the date filter rather than in post-processing.
What the search already answers
videoResults entries carry chapters on many rows as title and time pairs. That is a video's structure with no second call behind it, and it is the cheapest signal here for deciding what to open.
The video call adds lengthSeconds, category, keywords, isFamilySafe, isUnlisted, likes with extractedLikes, a description split into content and links, a channel block, captions, relatedVideos, endScreenVideos and an aiSummary.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago Changed · +6 lines 6d9985d0270a
- 23d ago First seen · 43 lines · 19 tokens per session scan A beab8667d40e
hasdata-youtube is a cursor rule published in the GitHub repository HasData/youtube-mcp (6 stars, last pushed 3d ago), licensed MIT. It adds 19 tokens to every session and 1,005 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-23.
Other cursor rules, from other repositories
bootstrap-alwayson
Minimal always-required LLM fixes.
code-optimization
Guidelines for optimizing duplicate and poorly structured code.
hasdata-booking
How to pull Booking.com hotel search and property details through the HasData MCP server.
hasdata-google-maps
How to pull Google Maps places, reviews, photos and posts through the HasData MCP server.
mcp-tools-readme
Keep README in sync when MCP tools change.
validate-design-early
Ask for design approval before implementing 100+ lines.