marketing-content

Router skill: maps a marketing-content request to exactly one specialist playbook. [DOC]

When to use

Producing or revising marketing artifacts: case studies, content calendars, copywriting, events, podcasts, press releases, video scripts, whitepapers. [DOC] NOT for: brand voice/tone defaults, internal comms, or sales decks — those are separate skills; routing them here loads the wrong playbook. [INFERENCE]

Inputs / outputs

• In: topic (one of the 8 enum values), depth (quick|deep). [CONFIG]
• Out: the artifact the resolved playbook specifies — never produced inline here. [DOC]

Procedure

1. Resolve topic from the request; if two topics plausibly fit, ask — do not guess (wrong route wastes the whole run). [INFERENCE]
2. Read EXACTLY ONE playbook from routes:. Never load the cluster. [CONFIG]
3. Execute that playbook along the spine: Discover → Analyze → Execute → Validate. [DOC]
4. depth=deep → apply exhaustively, verify each step; quick → essentials only. [CONFIG]

Topic routing (mirror of routes: — keep in sync)

case-study-writing · content-calendar · copywriting-frameworks · event-marketing · podcast-prep · press-release · video-script · whitepaper-creation [CONFIG]

Validation gate (before returning)

• Exactly one playbook was read; no sibling references loaded. [CONFIG]
• Output matches the resolved topic; depth honored. [DOC]
• Evidence tags present (Alfa set, see references/verification-tags.md); no invented prices; single-brand. [DOC]

Anti-patterns

• Routing on a keyword match without confirming intent. [INFERENCE]
• Producing the artifact in this file instead of delegating to the playbook. [DOC]
• Loading multiple playbooks "to compare" — defeats the router. [INFERENCE]

Quality gates: constitution v6.0.0 (enforcement), evidence tags, script-first rule. [DOC] Routing rubric + checklist: assets/ (quality-rubric.json, checklist.md). [CONFIG]