Install any skill in seconds. Free to start, no credit card required.
Get Started Free →This skill should be used when the user asks to "plan video content", "create a video calendar", "optimize video SEO", "analyze thumbnail performance", "improve video titles", "schedule video production", or "build a YouTube strategy".
.claude/skills/borghei-video-content-strategist/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -27% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -23% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 81% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 44% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 30% | 0% |
Production-ready video content strategy toolkit for planning content calendars, analyzing thumbnail effectiveness, and optimizing video metadata for platform SEO. Designed for content creators, marketing teams, and video producers managing consistent video output across YouTube, TikTok, LinkedIn, and other platforms.
Before planning, confirm these inputs. If any is unknown or vague, ASK — do not assume:
Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the artifact.
bash# Plan a video content calendar from topics and audience data python scripts/video_content_planner.py topics.json --weeks 8 --frequency 3 # Analyze thumbnail text and composition patterns python scripts/thumbnail_analyzer.py thumbnails.csv # Optimize video titles, descriptions, and tags for SEO python scripts/video_seo_optimizer.py video_data.json --platform youtube
| Tool | Purpose | Input | Output | |------|---------|-------|--------| | video_content_planner.py | Content calendar generation | JSON with topics/audience | Weekly calendar + production schedule | | thumbnail_analyzer.py | Thumbnail pattern analysis | CSV with thumbnail data | Optimization recommendations | | video_seo_optimizer.py | Video metadata SEO | JSON with video details | Optimized titles, descriptions, tags |
video_content_planner.py to generate 4-week calendarvideo_seo_optimizer.py for metadatathumbnail_analyzer.pyvideo_seo_optimizer.py on underperforming videos to identify metadata gapsthumbnail_analyzer.py on top vs bottom performersvideo_content_planner.py with --platforms youtube,tiktok,linkedinvideo_seo_optimizer.pySee references/video-strategy-guide.md for comprehensive frameworks covering:
json{ "channel": "TechStartupTV", "audience": { "primary": "SaaS founders, 25-45", "interests": ["startup growth", "fundraising", "product development"], "pain_points": ["scaling teams", "finding product-market fit", "managing burn rate"] }, "content_pillars": [ {"name": "Founder Stories", "ratio": 0.3, "format": "interview", "avg_length_min": 25}, {"name": "Tactical Guides", "ratio": 0.4, "format": "tutorial", "avg_length_min": 12}, {"name": "Industry Analysis", "ratio": 0.2, "format": "commentary", "avg_length_min": 8}, {"name": "Behind the Scenes", "ratio": 0.1, "format": "vlog", "avg_length_min": 5} ], "topics": [ {"title": "How We Hit $1M ARR", "pillar": "Founder Stories", "priority": "high"}, {"title": "5 Pricing Strategies That Work", "pillar": "Tactical Guides", "priority": "high"}, {"title": "AI in SaaS: 2026 Trends", "pillar": "Industry Analysis", "priority": "medium"} ] }
csvvideo_id,title,views,ctr_pct,has_face,has_text,text_words,colors_dominant,emotion V001,How to Scale,15000,8.2,yes,yes,3,red-yellow,surprise V002,Tech Review,8500,4.1,no,yes,5,blue-white,neutral
| Platform | Optimal Length | Max Recommended | |----------|---------------|-----------------| | YouTube (standard) | 8-15 min | 25 min | | YouTube Shorts | 30-60 sec | 60 sec | | TikTok | 30-90 sec | 3 min | | LinkedIn | 1-3 min | 10 min | | Instagram Reels | 15-60 sec | 90 sec |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | fail→fail | 16,250 | 16,083 | -1% | 1 | 1 | 0% | 2,232 | 3,549 | +59% | 0 | 0 | — |
case-01 | fail→fail | 13,871 | 11,763 | -15% | 1 | 1 | 0% | 2,073 | 3,160 | +52% | 0 | 0 | — |
case-02 | fail→pass | 15,791 | 5,103 | -68% | 1 | 1 | 0% | 2,769 | 2,017 | -27% | 0 | 0 | — |
case-03 | pass→pass | 17,188 | 14,863 | -14% | 1 | 1 | 0% | 2,744 | 3,866 | +41% | 0 | 0 | — |
case-05 | fail→pass | 13,520 | 2,924 | -78% | 1 | 1 | 0% | 2,027 | 1,565 | -23% | 0 | 0 | — |
case-06 | pass→pass | 11,887 | 8,814 | -26% | 1 | 1 | 0% | 1,732 | 2,608 | +51% | 0 | 0 | — |
case-07 | pass→fail | 11,079 | 12,579 | +14% | 1 | 1 | 0% | 1,714 | 3,167 | +85% | 0 | 0 | — |
case-08 | fail→pass | 9,769 | 9,893 | +1% | 1 | 1 | 0% | 1,571 | 2,838 | +81% | 0 | 0 | — |
case-09 | fail→pass | 8,569 | 5,176 | -40% | 1 | 1 | 0% | 1,377 | 1,987 | +44% | 0 | 0 | — |
case-10 | fail→pass | 8,205 | 2,590 | -68% | 1 | 1 | 0% | 1,260 | 1,644 | +30% | 0 | 0 | — |
case-11 | fail→pass | 6,860 | 1,645 | -76% | 1 | 1 | 0% | 996 | 1,467 | +47% | 0 | 0 | — |
case-12 | fail→pass | 7,943 | 1,661 | -79% | 1 | 1 | 0% | 1,197 | 1,502 | +25% | 0 | 0 | — |
case-22 | pass→pass | 8,199 | 11,748 | +43% | 1 | 1 | 0% | 1,437 | 3,499 | +143% | 0 | 0 | — |
case-13 | fail→pass | 10,065 | 5,573 | -45% | 1 | 1 | 0% | 1,557 | 1,817 | +17% | 0 | 0 | — |
case-14 | fail→pass | 10,975 | 6,393 | -42% | 1 | 1 | 0% | 2,248 | 2,201 | -2% | 0 | 0 | — |
case-15 | fail→pass | 14,378 | 11,421 | -21% | 1 | 1 | 0% | 2,388 | 3,179 | +33% | 0 | 0 | — |
case-16 | pass→pass | 12,305 | 8,264 | -33% | 1 | 1 | 0% | 2,185 | 2,556 | +17% | 0 | 0 | — |
case-17 | fail→pass | 7,724 | 3,329 | -57% | 1 | 1 | 0% | 1,075 | 1,748 | +63% | 0 | 0 | — |
case-18 | pass→pass | 8,213 | 1,853 | -77% | 1 | 1 | 0% | 1,279 | 1,525 | +19% | 0 | 0 | — |
case-19 | fail→pass | 3,766 | 1,667 | -56% | 1 | 1 | 0% | 487 | 1,463 | +200% | 0 | 0 | — |
case-20 | pass→pass | 14,852 | 28,225 | +90% | 1 | 1 | 0% | 2,230 | 3,921 | +76% | 0 | 0 | — |
case-21 | pass→pass | 12,383 | 11,233 | -9% | 1 | 1 | 0% | 1,898 | 2,976 | +57% | 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. The headline lift of +50 percentage points is the difference between those two pass rates over the 22 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.