Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate comprehensive content briefs for writers, ensuring clarity, alignment, and strategic content creation across all formats.
.claude/skills/nicepkg-content-brief-generator/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-10 | ✗→✓ | ▲ Improved | 145% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 110% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 146% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 101% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 137% | 0% |
A comprehensive skill for creating detailed content briefs that guide writers, ensure consistency, and align content with business goals.
Helps you create professional content briefs for:
Without a brief:
With a brief:
Use for: Thought leadership, tutorials, listicles, how-tos
Key elements:
Use for: API docs, guides, developer documentation
Key elements:
Use for: Product pages, signup pages, campaign pages
Key elements:
Use for: Customer success stories, testimonials
Key elements:
Best for: Marketing copy, sales pages
Best for: Blog posts, email campaigns
Best for: News, technical docs
Best for: Case studies, narratives
Define your brand voice:
Use generate_brief.sh to create comprehensive brief
Writer researches and creates outline
Writer creates content following brief
Editor reviews against brief requirements
SEO review and optimization
Stakeholder sign-off
Publish and distribute
Monitor metrics defined in brief
Too Vague: "Write about our product" → No clear audience, goal, or angle
No Research: Missing competitor analysis and keyword research → Content won't rank or differentiate
Unclear Voice: "Make it engaging" → Not specific enough for consistency
Missing CTAs: No clear next step for readers → Missed conversion opportunities
Ignoring SEO: No keyword strategy → Content won't be discovered
Specific: "Write 1500-word tutorial for junior developers on JWT authentication, including code examples in Node.js"
Research-Backed: "Target keyword 'JWT authentication tutorial' (2.4K monthly searches), compete with existing articles from Auth0 and FreeCodeCamp"
Clear Voice: "Technical but approachable - explain like a senior dev mentoring a junior, use examples, avoid jargon"
Action-Oriented: "CTA: Sign up for free API key, secondary: Join Discord community"
SEO-Optimized: "Primary: 'JWT authentication', Secondary: 'JSON web token Node.js', 'secure API authentication'"
bash./scripts/generate_brief.sh
Interactive workflow will guide you through:
bash./scripts/validate_brief.sh path/to/brief.md
Checks for:
references/content_brief_template.md - Complete template
references/seo_guidelines.md - SEO best practices
references/content_frameworks.md - Structure frameworks
references/tone_voice_guide.md - Voice guidelinesSystem generates comprehensive brief with:
System creates technical brief with:
System builds conversion-focused brief with:
Brief creation triggers calendar entry
Keywords from SEMrush, Ahrefs, etc.
Brief becomes Jira/Asana task
Brief exported to content management system
Success metrics tracked in dashboards
All reference materials included:
Content briefs ensure:
Use this skill to create comprehensive briefs that guide writers to success.
Remember: Time spent on a good brief saves 10x time in revisions. Get it right before writing begins.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-10 | fail→pass | 15,089 | 17,622 | +17% | 1 | 1 | 0% | 2,171 | 5,311 | +145% | 0 | 0 | — |
case-01 | pass→pass | 12,619 | 14,745 | +17% | 1 | 1 | 0% | 2,036 | 4,562 | +124% | 0 | 0 | — |
case-02 | pass→pass | 12,210 | 8,585 | -30% | 1 | 1 | 0% | 1,904 | 3,953 | +108% | 0 | 0 | — |
case-03 | fail→pass | 12,123 | 8,720 | -28% | 1 | 1 | 0% | 1,901 | 3,995 | +110% | 0 | 0 | — |
case-04 | fail→pass | 13,141 | 13,261 | +1% | 1 | 1 | 0% | 1,947 | 4,785 | +146% | 0 | 0 | — |
case-05 | fail→pass | 13,524 | 9,200 | -32% | 1 | 1 | 0% | 2,072 | 4,162 | +101% | 0 | 0 | — |
case-06 | pass→pass | 10,542 | 9,265 | -12% | 1 | 1 | 0% | 1,675 | 4,185 | +150% | 0 | 0 | — |
case-07 | pass→pass | 13,201 | 15,879 | +20% | 1 | 1 | 0% | 2,061 | 4,894 | +137% | 0 | 0 | — |
case-08 | pass→pass | 14,549 | 13,749 | -5% | 1 | 1 | 0% | 2,123 | 4,730 | +123% | 0 | 0 | — |
case-09 | fail→fail | 15,727 | 17,783 | +13% | 1 | 1 | 0% | 2,134 | 5,210 | +144% | 0 | 0 | — |
case-11 | fail→pass | 7,975 | 2,737 | -66% | 1 | 1 | 0% | 1,303 | 3,094 | +137% | 0 | 0 | — |
case-12 | fail→pass | 8,604 | 1,916 | -78% | 1 | 1 | 0% | 1,254 | 2,920 | +133% | 0 | 0 | — |
case-13 | pass→pass | 13,701 | 11,954 | -13% | 1 | 1 | 0% | 1,906 | 4,254 | +123% | 0 | 0 | — |
case-14 | pass→pass | 10,295 | 2,376 | -77% | 1 | 1 | 0% | 1,480 | 2,985 | +102% | 0 | 0 | — |
case-20 | fail→pass | 6,215 | 1,621 | -74% | 1 | 1 | 0% | 897 | 2,853 | +218% | 0 | 0 | — |
case-15 | pass→pass | 15,102 | 12,366 | -18% | 1 | 1 | 0% | 2,072 | 4,397 | +112% | 0 | 0 | — |
case-16 | fail→pass | 10,173 | 4,679 | -54% | 1 | 1 | 0% | 1,531 | 3,269 | +114% | 0 | 0 | — |
case-17 | pass→pass | 15,689 | 13,699 | -13% | 1 | 1 | 0% | 2,217 | 4,730 | +113% | 0 | 0 | — |
case-18 | fail→pass | 12,118 | 13,803 | +14% | 1 | 1 | 0% | 1,844 | 4,730 | +157% | 0 | 0 | — |
case-19 | fail→pass | 8,774 | 2,223 | -75% | 1 | 1 | 0% | 1,285 | 3,000 | +133% | 0 | 0 | — |
case-21 | fail→fail | 4,650 | 2,792 | -40% | 1 | 1 | 0% | 698 | 3,010 | +331% | 0 | 0 | — |
case-22 | pass→pass | 12,486 | 9,459 | -24% | 1 | 1 | 0% | 1,872 | 4,058 | +117% | 0 | 0 | — |
case-23 | pass→pass | 17,072 | 14,998 | -12% | 1 | 1 | 0% | 3,240 | 5,640 | +74% | 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. 23 cases were attempted. The headline lift of +43 percentage points is the difference between those two pass rates over the 23 comparable cases.
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.