Playbook Builder

Generate authority-establishing, interactive single-page HTML playbooks for any industry. One command produces a fully deployed, lead-generating playbook with self-assessments, AI prompts, readiness checklists, and maturity models — backed by 30+ minutes of deep research.

5-Phase Workflow

1. Research — 3 parallel agents (Pain Hunter, Stat Collector, Trend Scanner) gather 30+ substantiated data points
2. Content — Generate 13-section playbook using research output (problem-first, no generic advice, real numbers only)
3. HTML — Build interactive single-page HTML matching the notebook/paper template design system
4. Analytics — Inject tracking script (or no-op stub if Supabase not configured)
5. Deploy — Human review gate, then Vercel deployment + registry update

Quick Interview (Ask Before Building)

1. Industry/domain: "What industry is this playbook for?" (e.g., pharmacy chains, dental clinics)
2. Target audience: "Who reads this? What's their title?" (e.g., chain owners, clinic directors)
3. Geography: "India-focused, or global?" (affects stats, compliance, currency)
4. Insider knowledge: "Any specific pain points, stats, or insights from this industry?" (optional)
5. Branding: "JICATE Solutions branding, or white-label for a client?" — If white-label: client name, primary color hex, CTA email/URL

Pre-Flight Checks

1. Concurrency guard: Read /Users/omm/PROJECTS/jicate-playbooks/registry.json — abort if same industry slug has "status": "in-progress"
2. Industry scoping: If industry is a top-level category (healthcare, manufacturing, education, retail, agriculture, finance, logistics, construction, hospitality, media), ask user to narrow to a segment
3. Firecrawl credits: Run firecrawl --status to verify sufficient credits before committing to 30+ minutes of research
4. Bootstrap: If /Users/omm/PROJECTS/jicate-playbooks/ does not exist, create it with empty registry.json and registry-config.json

Knowledge Graph

Start with references/phase-1-research.md — it defines the 3 parallel research agents, fallback chains, and the research output schema that feeds all subsequent phases.

When generating section content, read references/phase-2-content.md for the 13-section structure, content quality rules, and how to reference JICATE OS capabilities and the case study.

When building the HTML file, read references/phase-3-html.md for the template structure, interactive element specs, and white-label sanitization rules. Also read references/design-system.md for exact colors, fonts, and component styles.

When injecting analytics, read references/phase-4-analytics.md for the tracking script, Supabase schema, Edge Function security, and privacy/consent requirements.

When deploying and packaging, read references/phase-5-deploy.md for the human review gate, Vercel deployment, registry management, and paid client deliverables.

Before marking any phase complete, read references/quality-gates.md for the 9 quality gates, banned phrases, retry limits, and failure handling.

For SEO metadata and social sharing tags, read references/seo-metadata.md before finalizing the HTML.

For --update invocations, read references/phase-5-deploy.md for version management — old version archived, new version deployed to same URL.