Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Query YouTube Data API v3 — channel stats, recent videos, top videos, comments. Supports multi-account (OAuth or API Key). Use when user asks about YouTube metrics, YouTube channel, subscribers, views, videos, engagement, or any reference to YouTube analytics.
.claude/skills/evolution-foundation-int-youtube/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | -42% | 0% |
| case-11 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-15 | ✗→✓ | ▲ Improved | -32% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 21% | 0% |
| case-17 | ✗→✓ | ▲ Improved | -38% | 0% |
YouTube integration to monitor Evolution channels and others. Supports multiple accounts via OAuth (Social Auth App) or API Key.
Accounts configured via make social-auth (OAuth login) or manually in .env:
envSOCIAL_YOUTUBE_1_LABEL=Evolution API SOCIAL_YOUTUBE_1_ACCESS_TOKEN=ya29... SOCIAL_YOUTUBE_1_CHANNEL_ID=UC9kZHm3TnEt41ztGOLyQO9g SOCIAL_YOUTUBE_1_REFRESH_TOKEN=1//0h...
bashpython3 {project-root}/.claude/skills/int-youtube/scripts/youtube_client.py <command> [args]
bash# List configured accounts youtube_client.py accounts # Channel stats (subscribers, views, total videos) youtube_client.py channel_stats [account_label_or_index] # Last N videos with metrics (via playlistItems — 3 units) youtube_client.py recent_videos [account] [N] # Top N videos by views youtube_client.py top_videos [account] [N] # Stats for specific videos youtube_client.py video_stats VIDEO_ID [VIDEO_ID...] # Comments on a video youtube_client.py comments VIDEO_ID [N] # Summary of all accounts youtube_client.py summary
json{ "account": "Evolution API", "channel_id": "UC9kZHm3TnEt41ztGOLyQO9g", "subscribers": 7450, "total_views": 132462, "video_count": 27, "videos": [ { "id": "abc", "title": "...", "published": "2026-...", "views": 7180, "likes": 500, "comments": 164, "engagement_rate": 9.25, "url": "https://youtube.com/watch?v=abc" } ] }
playlistItems: 1 unit (used instead of search which costs 100)channels, videos, commentThreads: 1 unit each| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 3,603 | 7,727 | +114% | 1 | 1 | 0% | 507 | 1,255 | +148% | 0 | 0 | — |
case-02 | fail→fail | 7,809 | 10,007 | +28% | 1 | 1 | 0% | 1,370 | 1,536 | +12% | 0 | 0 | — |
case-03 | fail→fail | 9,021 | 6,880 | -24% | 1 | 1 | 0% | 1,444 | 889 | -38% | 0 | 0 | — |
case-04 | pass→pass | 13,987 | 14,394 | +3% | 1 | 1 | 0% | 2,677 | 2,884 | +8% | 0 | 0 | — |
case-05 | pass→fail | 8,904 | 5,810 | -35% | 1 | 1 | 0% | 1,579 | 928 | -41% | 0 | 0 | — |
case-06 | fail→fail | 2,969 | 8,582 | +189% | 1 | 1 | 0% | 417 | 1,317 | +216% | 0 | 0 | — |
case-07 | pass→pass | 11,825 | 6,483 | -45% | 1 | 1 | 0% | 2,039 | 1,637 | -20% | 0 | 0 | — |
case-08 | pass→pass | 16,733 | 6,774 | -60% | 1 | 1 | 0% | 2,152 | 1,737 | -19% | 0 | 0 | — |
case-09 | fail→fail | 8,164 | 2,131 | -74% | 1 | 1 | 0% | 1,401 | 981 | -30% | 0 | 0 | — |
case-10 | fail→pass | 8,377 | 1,604 | -81% | 1 | 1 | 0% | 1,526 | 888 | -42% | 0 | 0 | — |
case-11 | fail→pass | 8,241 | 1,757 | -79% | 1 | 1 | 0% | 1,372 | 934 | -32% | 0 | 0 | — |
case-12 | pass→pass | 5,981 | 1,647 | -72% | 1 | 1 | 0% | 1,085 | 810 | -25% | 0 | 0 | — |
case-13 | fail→fail | 8,418 | 2,062 | -76% | 1 | 1 | 0% | 1,531 | 963 | -37% | 0 | 0 | — |
case-14 | fail→fail | 6,259 | 1,650 | -74% | 1 | 1 | 0% | 1,334 | 900 | -33% | 0 | 0 | — |
case-15 | fail→pass | 7,498 | 1,194 | -84% | 1 | 1 | 0% | 1,119 | 760 | -32% | 0 | 0 | — |
case-16 | fail→pass | 4,238 | 1,342 | -68% | 1 | 1 | 0% | 690 | 836 | +21% | 0 | 0 | — |
case-17 | fail→pass | 8,457 | 1,178 | -86% | 1 | 1 | 0% | 1,365 | 850 | -38% | 0 | 0 | — |
case-18 | pass→pass | 6,717 | 1,751 | -74% | 1 | 1 | 0% | 1,368 | 959 | -30% | 0 | 0 | — |
case-19 | pass→pass | 8,887 | 4,529 | -49% | 1 | 1 | 0% | 1,669 | 1,386 | -17% | 0 | 0 | — |
case-20 | pass→pass | 11,982 | 2,669 | -78% | 1 | 1 | 0% | 2,120 | 1,138 | -46% | 0 | 0 | — |
case-21 | fail→pass | 10,991 | 1,400 | -87% | 1 | 1 | 0% | 1,834 | 848 | -54% | 0 | 0 | — |
case-22 | fail→pass | 6,227 | 2,081 | -67% | 1 | 1 | 0% | 963 | 988 | +3% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 17 counted toward the lift figure. The other 5 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +27 percentage points is the difference between those two pass rates over the 17 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.