creo
Comprehensive design, UX, content, DevOps, and testing toolkit for web applications. Design reviews, UX flow analysis, marketing content, SEO auditing, CI/CD pipelines, test orchestration, performance
SKILL.mdcreo/SKILL.md
GitHubCreo -- Design & Development Toolkit
Creo is a comprehensive design, UX, content, DevOps, and testing toolkit for web applications. It provides a unified interface for design reviews, UX flow analysis, marketing content generation, SEO auditing, CI/CD pipeline management, test orchestration, and full marketing site creation. Each command routes to a specialized sub-skill or subagent optimized for its domain.
Quick Reference
| Command | Description |
|---|---|
/creo design-review <url> |
UI/UX review (responsive, WCAG AA, Nielsen's heuristics) |
/creo design-implement <report> |
Execute design fixes from review reports |
/creo ux-internal <flow> |
Analyze your own app's UX flows |
/creo ux-competitor <url> |
Analyze competitor websites |
/creo content <page-type> |
Generate marketing copy (JTBD, pain points) |
/creo image-prompt <context> |
Generate image prompts for DALL-E/Midjourney/SDXL |
/creo seo <url> |
Next.js SEO + GEO audit (7-phase, technical + schema + GEO) |
/creo seo onboard <url> |
Full SEO lifecycle: audit, plan, guided GSC setup, monitoring cadence |
/creo seo init |
Scan codebase and cache project profile for faster audits |
/creo seo autofix |
Closed-loop GSC remediation (allowlisted fixes, ledger, live verification) |
/creo seo semantic-core |
Focused query core with noise filtering and P0-P3 priorities |
/creo seo weekly |
Weekly rank history + LLM-citability monitoring |
/creo seo-content <url> |
Deep content audit (E-E-A-T, AI-pattern detection, citability) |
/creo devops <command> |
Infrastructure management (GitHub/Cloudflare/Railway/Stripe) |
/creo pipeline <command> |
CI/CD pipeline specialist (GitHub Actions) |
/creo test <command> |
Test orchestration (unit + E2E) |
/creo marketing-site <command> |
Full marketing site creation (7-stage) |
/creo ai-generation <command> |
AI generation pipeline expertise |
/creo perf <command> |
Performance optimization: init, baseline, audit, optimize, report (six-layer, measured) |
Orchestration Logic
Individual Commands
For any single command (e.g., /creo design-review, /creo seo), the orchestrator loads the relevant sub-skill directly. The sub-skill file contains the full instructions and references needed for that domain.
Marketing Site Full Build (/creo marketing-site full)
This is a multi-stage orchestration that delegates to parallel subagents:
- Stage 1-2: Infrastructure setup and content generation run sequentially.
- Stage 3: Page component assembly.
- Stage 4-5: Parallel delegation to:
creo-content-strategist-- generates marketing copy for all pagescreo-seo-- runs codebase audit and optimizes metadata/structured datacreo-design-review-- validates responsive design and accessibilitycreo-ux-competitor-- analyzes competitor sites for differentiation
- Stage 6: Localization via
creo-i18n-translator. - Stage 7: Final QA combining all subagent reports.
DevOps Routing (/creo devops)
Routes to specialized CLI subagents based on the target platform:
| Subcommand | Subagent |
|---|---|
devops github |
creo-github-cli |
devops cloudflare |
creo-cloudflare-cli |
devops railway |
creo-railway-cli |
devops stripe |
creo-stripe-cli |
Performance Routing (/creo perf)
Loads the creo-perf sub-skill. perf audit spawns the measure-only creo-perf-audit subagent; perf init scaffolds the project's .claude/skills/creo-perf/ via the perf-harness extension; perf optimize runs the fix loop in the main context (baseline, one concern, re-measure, verify, commit).
Test Routing (/creo test)
Routes to the appropriate testing subagent:
| Subcommand | Subagent |
|---|---|
test unit |
creo-unit-test |
test e2e |
creo-e2e-test |
test all |
Both subagents run in parallel |
Domains
| Domain | Sub-Skills | Primary Focus |
|---|---|---|
| Design & UX | design-review, design-implement, ux-internal, ux-competitor | Visual quality, accessibility, usability |
| Content & Marketing | content, image-prompt | Copy generation, JTBD frameworks, image prompts |
| SEO | seo | Technical SEO, structured data, Core Web Vitals |
| DevOps | devops (github, cloudflare, railway, stripe) | Infrastructure, deployment, services |
| Testing | test (unit, e2e) | Code quality, test coverage, E2E validation |
| Performance | perf | Six-layer audit, immutable baselines, before/after proof, fix loop |
| Orchestration | marketing-site, ai-generation | Multi-stage workflows, parallel delegation |
Reference Files
On-demand reference files loaded by sub-skills as needed. Located in creo/references/:
| File | Used By | Content |
|---|---|---|
design-principles.md |
design-review, design-implement | S-tier SaaS design checklist |
design-style-guide.md |
design-review, design-implement | Viewports, spacing, typography, accessibility |
visual-dev-snippets.md |
design-review, design-implement | Quick visual verification workflow |
ux-methodologies.md |
ux-internal, ux-competitor | Nielsen's heuristics, JTBD, journey mapping |
ux-report-template.md |
ux-internal, ux-competitor | Structured UX analysis report format |
pain-points-framework.md |
content, marketing-site | Customer pain analysis by segment |
voice-and-tone.md |
content, marketing-site | Brand voice pillars, writing principles |
image-style-reference.md |
image-prompt | Prompt patterns, style modifiers, categories |
seo-checklist.md |
seo, marketing-site | SEO and i18n implementation checklist |
seo-audit-checklists.md |
seo | Phase-based SEO audit (5 phases) |
i18n-translation-guide.md |
marketing-site, content | Localization best practices, language list |
workflow-stages.md |
marketing-site | 7-stage marketing site creation workflow |
Sub-Skills
Each sub-skill is a standalone SKILL.md that can be invoked directly or through the orchestrator.
| # | Sub-Skill | Path | Trigger |
|---|---|---|---|
| 1 | Design Review | creo/skills/design-review/SKILL.md |
/creo design-review |
| 2 | Design Implementation | creo/skills/design-implement/SKILL.md |
/creo design-implement |
| 3 | UX Internal Analysis | creo/skills/ux-internal/SKILL.md |
/creo ux-internal |
| 4 | UX Competitor Analysis | creo/skills/ux-competitor/SKILL.md |
/creo ux-competitor |
| 5 | Content Strategy | creo/skills/content/SKILL.md |
/creo content |
| 6 | Image Prompt Generation | creo/skills/image-prompt/SKILL.md |
/creo image-prompt |
| 7 | SEO Audit | creo/skills/seo/SKILL.md |
/creo seo |
| 8 | DevOps Management | creo/skills/devops/SKILL.md |
/creo devops |
| 9 | CI/CD Pipeline | creo/skills/pipeline/SKILL.md |
/creo pipeline |
| 10 | Test Orchestration | creo/skills/test/SKILL.md |
/creo test |
| 11 | Marketing Site | creo/skills/marketing-site/SKILL.md |
/creo marketing-site |
| 12 | AI Generation | creo/skills/ai-generation/SKILL.md |
/creo ai-generation |
| 13 | Performance | skills/creo-perf/SKILL.md |
/creo perf |
Subagents
Each subagent is a focused agent definition for parallel delegation.
| # | Subagent | Path | Role |
|---|---|---|---|
| 1 | creo-design-review | creo/agents/design-review.md |
Visual review and accessibility |
| 2 | creo-design-implement | creo/agents/design-implement.md |
Execute design fixes |
| 3 | creo-ux-internal | creo/agents/ux-internal.md |
Internal UX flow analysis |
| 4 | creo-ux-competitor | creo/agents/ux-competitor.md |
Competitor UX analysis |
| 5 | creo-content-strategist | creo/agents/content-strategist.md |
Marketing copy generation |
| 6 | creo-image-prompt | creo/agents/image-prompt.md |
Image prompt generation |
| 7 | creo-seo | agents/creo-seo.md |
Next.js SEO + GEO audit (7-phase) |
| 7b | creo-seo-content | agents/creo-seo-content.md |
Deep content-quality audit (delegate or standalone) |
| 8 | creo-github-cli | creo/agents/github-cli.md |
GitHub API operations |
| 9 | creo-cloudflare-cli | creo/agents/cloudflare-cli.md |
Cloudflare management |
| 10 | creo-railway-cli | creo/agents/railway-cli.md |
Railway deployment |
| 11 | creo-stripe-cli | creo/agents/stripe-cli.md |
Stripe integration |
| 12 | creo-e2e-test | creo/agents/e2e-test.md |
End-to-end testing |
| 13 | creo-perf-audit | agents/creo-perf-audit.md |
Six-layer performance audit (measure-only) |
Optional Extensions
These extensions can be added to the toolkit as needed:
- image-generation -- Integration with DALL-E, Midjourney, or SDXL APIs for automated image creation from prompts generated by the image-prompt sub-skill.
- i18n-translator -- Automated translation pipeline for marketing site localization across 39+ languages with quality validation.
- gsc-analyzer -- Google Search Console analysis, SEO auditing, and operations automation: 17 analyzers, GSC API + Indexing API clients, UI-only surface automation (drilldown exports, Validate Fix), the closed-loop autofix scripts, weekly rank/LLM-citability monitoring, and a full GSC API reference.
- perf-harness -- Deterministic benchmark scripts behind
/creo perf: scenario captures (EXPLAIN medians, p50/p95 loops, wire bytes, ETag replay, Playwright interaction probes), Lighthouse initial-load audit, platform/schema audits, pg_stat_statements workload discovery, auto-built before/after dashboard, and the/creo perf initproject scaffolder.
Project Configuration
All skills check .claude/project-config.md at runtime for project-specific settings. This file should contain:
- Dev server URL -- The local development server address (default:
http://localhost:3000). - Framework -- The frontend framework in use (Next.js, Vite, etc.).
- CSS framework -- Tailwind CSS, MUI, shadcn/ui, or custom.
- Supported locales -- List of language codes for i18n.
- Brand guidelines -- Brand name, colors, fonts, voice attributes.
- Deployment targets -- Hosting platform and CI/CD configuration.
- Test framework -- Vitest, Jest, Playwright, Cypress, etc.
If .claude/project-config.md does not exist, skills use sensible defaults and prompt the user for required configuration.