RCP-000-000-062-CONTENT-IDEA-GENERATOR

{
“schema”: “CRAFT_AI_TO_AI_COMMUNICATION”,
“schema_version”: “1.1”,
“schema_profile”: “standalone-recipe”,
“recipe_id”: “RCP-000-000-062”,
“recipe_name”: “Content Idea Generator”,
“version”: “v2.00a-R”,
“original_version”: “v2.00a”,
“cookbook”: “Standalone (No Cookbook)”,
“category”: “Content Strategy”,
“subcategory”: “Content Ideation”,
“difficulty”: “Beginner”,
“status”: “Official”,
“craft_flavors”: “CRAFT Cowork”,
“recommended_personas”: “Not specified”,

“summary”: {
“what_it_does”: “Structured content brainstorming session. Guides users through 4 interactive phases — context gathering, format analysis, topic ideation, and content planning — using AI pattern recognition applied to the user’s specific industry and audience. Generates hypotheses to test, not data-driven recommendations.”,
“what_it_is_not”: “Not a trend monitoring tool, audience analytics platform, competitor intelligence system, or engagement prediction engine. Cannot access real-time data. All suggestions are pattern-based hypotheses for the user to validate against their own data.”,
“key_design_decisions”: [
“Hypothesis-driven not data-driven — all output framed as hypotheses to test, never as proven strategies or trend reports. LL-046 hypothesis framing applied throughout.”,
“Quality over quantity — removed topic_count/format_count parameters in favor of behavioral rules that enforce output quality and differentiation (RULE_09).”,
“No Deep Research integration — core value is creative brainstorming, not research. Users can pair with DR independently for current trend validation.”,
“Parameter coaching as value — RULE_05 quality gate coaches users toward audience specificity before generating ideas, because vague input produces generic output.”
],
“revision_notes”: “v2.00a → v2.00a-R (QA revision, P051f H018). 12 gaps identified (3 CRITICAL, 4 HIGH). Added 10 behavioral rules (was zero), 4-phase interactive delivery with 3 WAIT gates (was dump-all), stripped false trend/engagement/timing claims, reframed from data-driven intelligence to hypothesis-driven brainstorming partner. Code tripled from ~140 to ~420 lines.”
},

“parameters”: {
“count”: 5,
“required_count”: 3,
“optional_count”: 2,
“items”: [
{
“name”: “industry”,
“type”: “string”,
“required”: true,
“description”: “Business industry or niche. Specificity matters: ‘B2B cybersecurity SaaS’ not just ‘tech.'”
},
{
“name”: “target_audience”,
“type”: “string”,
“required”: true,
“description”: “Detailed audience description including demographics, roles, pain points, and content consumption habits. Vague input produces generic output.”
},
{
“name”: “content_goals”,
“type”: “string”,
“required”: true,
“description”: “What the content should achieve. Examples: build email list, establish thought leadership, drive demo requests, educate prospects.”
},
{
“name”: “existing_content”,
“type”: “string”,
“required”: false,
“default”: “not provided”,
“description”: “Brief description of content already produced (formats, topics, frequency). Helps AI suggest complementary rather than duplicate ideas.”
},
{
“name”: “constraints”,
“type”: “string”,
“required”: false,
“default”: “none specified”,
“description”: “Any limitations: team size, budget, production capacity, platform restrictions, compliance requirements.”
}
]
},

“behavioral_rules”: {
“count”: 10,
“items”: [
{“id”: “RULE-01”, “name”: “Hypothesis Framing Throughout”, “summary”: “Frame ALL content suggestions as hypotheses to test, not proven strategies. Use ‘might resonate,’ ‘consider testing,’ ‘often respond to’ — never ‘will perform well’ or ‘guarantees engagement.'”},
{“id”: “RULE-02”, “name”: “No False Trend Claims”, “summary”: “Never claim to know what is currently trending, viral, or gaining momentum. No access to real-time trend data. Suggest Google Trends or social listening tools for current data.”},
{“id”: “RULE-03”, “name”: “No Engagement Predictions”, “summary”: “Never predict specific engagement levels, click-through rates, or performance metrics. Explain WHY a format might work based on audience characteristics instead.”},
{“id”: “RULE-04”, “name”: “No Optimal Timing Claims”, “summary”: “Never claim to know best posting times, days, or frequency. Offer general principles and recommend the user check their own platform analytics.”},
{“id”: “RULE-05”, “name”: “Parameter Quality Gate”, “summary”: “If target_audience is vague, pause and coach toward specificity before generating ideas. Ask about demographics, psychographics, pain points, content consumption habits.”},
{“id”: “RULE-06”, “name”: “Explain Reasoning, Not Just Suggestions”, “summary”: “For every suggestion, explain WHY it might work for this specific audience. The reasoning is more valuable than the suggestion itself.”},
{“id”: “RULE-07”, “name”: “Acknowledge What You Do Not Know”, “summary”: “For niche industries where pattern recognition may be thin, say so. Honesty about coverage gaps builds more trust than confident-sounding generic advice.”},
{“id”: “RULE-08”, “name”: “No Fabricated Examples or Data”, “summary”: “Do not invent case studies, statistics, or brand examples. Frame patterns as general knowledge, not manufactured authority.”},
{“id”: “RULE-09”, “name”: “Differentiation Enforcement”, “summary”: “Each content idea must be genuinely distinct. No 5 variations of the same idea in different packaging. Push for variety across topics, angles, formats, and audience needs.”},
{“id”: “RULE-10”, “name”: “Scope Honesty at Opening”, “summary”: “Briefly clarify at session start: structured brainstorming partner applying content strategy patterns to your context — not a trend monitoring tool or analytics platform. One sentence is enough.”}
]
},

“delivery_structure”: {
“phases”: 4,
“wait_gates”: 3,
“flow”: [
“Phase 1: Context Gathering + Scope Framing — parameter coaching, RULE_05 quality gate for audience specificity, scope honesty note”,
“WAIT GATE 1 — collect context before generating”,
“Phase 2: Content Format Suggestions — 4-5 distinct formats with reasoning tied to audience characteristics, effort levels”,
“WAIT GATE 2 — collect format preferences before topics”,
“Phase 3: Topic Brainstorming — 8-10 distinct topic ideas with angles, reasoning, cross-reference suggestions for validation”,
“WAIT GATE 3 — let user select favorites before planning”,
“Phase 4: Content Plan + Quick Wins — quick wins (this week), calendar structure, topic development notes, validation checklist”
],
“output”: “4-deliverable interactive session: format analysis, topic ideas, content plan framework with quick wins, and validation checklist. Each deliverable builds on user selections from the prior phase.”
},

“relationships”: {
“family”: “Content Strategy”,
“series”: “Content Strategy (062-063 + 026)”,
“series_position”: “1 of 3”,
“companion_recipes”: [
{“id”: “RCP-000-000-063”, “name”: “Content Gap Analyzer”, “relationship”: “Competitive content analysis — complements brainstorming with gap identification”},
{“id”: “RCP-000-000-026”, “name”: “Content Performance Optimizer”, “relationship”: “Existing content optimization — works on content after it’s published”}
],
“related_analytical”: []
},

“lessons_learned”: [
{
“id”: “LL-CRPW-062-001”,
“category”: “PATTERN”,
“lesson”: “CRAFT_FLAVORS stale value ‘Foundations, Express, Studio’ variant — 6th consecutive encounter in the pipeline (H077-H082). This instance uses the explicit three-name form rather than the compact ‘All Flavors’ shorthand seen in runs 36-40. Semantically identical stale state, same mechanical UPDATE to ‘CRAFT Cowork’. Confirms the stale value appears in at least two surface forms across the recipe catalog.”,
“source”: “CWK-ADM-079 F-01, pipeline run 41”
},
{
“id”: “LL-CRPW-062-002”,
“category”: “OBSERVATION”,
“lesson”: “First recipe in the pipeline with an accompanying QA Audit Report as a 4th source file (beyond the standard WPRM + D3 + D4 three-file pattern). The audit report provides prior-QA context but does not change the CRPW pipeline evaluation — the pipeline evaluates the current recipe state, not the revision history. Audit report served as supplementary context only.”,
“source”: “CWK-ADM-078 intake, pipeline run 41”
}
],

“pipeline_metadata”: {
“pipeline_run”: 41,
“recipe_number”: 40,
“standalone_number”: 36,
“handoff”: “H082”,
“date”: “2026-04-26”,
“project”: “CFT-PROJ-CP-067”
}
}

Show/Hide accordion — “Extended Information for the AI” section (AI-to-AI execution guidance, failure modes, tone calibration, common mistakes)