# daily-blog-production > End-to-end workflow for producing one blog post per day, with synthesis coordination, quality gates, and social distribution - Author: Aether - Repository: coreycottrell/aiciv-jared - Version: 20260129234749 - Stars: 0 - Forks: 0 - Last Updated: 2026-02-07 - Source: https://github.com/coreycottrell/aiciv-jared - Web: https://mule.run/skillshub/@@coreycottrell/aiciv-jared~daily-blog-production:20260129234749 --- --- name: daily-blog-production description: End-to-end workflow for producing one blog post per day, with synthesis coordination, quality gates, and social distribution --- # Daily Blog Production SKILL **Purpose**: Produce one high-quality, opinionated blog post every day with coordinated social distribution. **Owner**: doc-synthesizer (writing) + the-conductor (orchestration) **Created**: 2026-01-02 **Status**: DESIGN DOCUMENT --- ## Quick Reference ### Daily Production Overview | Phase | Duration | Owner | Output | |-------|----------|-------|--------| | Morning Research | 15-20 min | web-researcher | Research brief | | Synthesis & Writing | 30-45 min | doc-synthesizer | Draft post | | Quality Gate | 10 min | doc-synthesizer | Final post | | Social Prep | 15 min | doc-synthesizer | Thread + LinkedIn | | Distribution | 5 min | the-conductor | Posted + tracked | | **Total** | **75-95 min** | - | Complete package | --- ## 1. Input Requirements from Research ### What I Need The morning research brief should deliver: ```markdown ## Research Brief: [Topic] **Date**: YYYY-MM-DD **Researcher**: web-researcher ### Core Finding [1-2 sentences: What's the headline insight?] ### Supporting Evidence - [Fact 1 with source] - [Fact 2 with source] - [Fact 3 with source] ### CEO vs Employee Angle [If applicable: How does this relate to AI taking work vs humans directing AI?] ### Contrarian Take [What's the non-obvious perspective? What's everyone getting wrong?] ### AI-CIV Connection [How does this connect to what we're building? Is there a teaching moment?] ### Sources - [URL 1] - [URL 2] - [URL 3] ``` ### What I DON'T Need - Comprehensive literature reviews - All possible perspectives - Academic citations beyond essentials - More than 500 words of input **Principle**: I synthesize better from focused, opinionated briefs than from exhaustive dumps. The researcher's job is to have a point of view. --- ## 2. Daily Post vs Weekly Deep-Dive Structure ### Daily Posts (800-1200 words, 5x/week) **Goal**: React fast, be useful, stay authentic ```markdown # [Punchy Title - States a Position] *[Brief intro - why this matters today]* --- ## The Setup [What happened / what's the context - 150 words max] --- ## What Most People Are Missing [The contrarian insight - this is the core value - 300 words] --- ## The CEO vs Employee Lens [If relevant: Frame through the "who directs vs who executes" question - 200 words] --- ## What This Means for You [Practical implications - NOT generic advice - 200 words] --- ## What We're Doing About It [AI-CIV connection - how we're exploring/building - 150 words] --- *Written by [agent] on behalf of ${CIV_NAME}* *[Date]* ``` ### Weekly Deep-Dives (1500-2500 words, 1x/week - Saturdays) **Goal**: Go deeper, establish authority, create reference material ```markdown # [Comprehensive Title] *[Longer intro with stakes clearly stated]* --- ## Executive Summary [For skimmers - 200 words] --- ## Part 1: The Landscape [Comprehensive context - 400 words] --- ## Part 2: The Analysis [Deep dive into implications - 600 words] --- ## Part 3: The Contrarian Position [Our differentiated take - 400 words] --- ## Part 4: Practical Framework [Tools, frameworks, actionable approaches - 400 words] --- ## Part 5: Where This Goes [Future implications, predictions - 300 words] --- *[Standard footer]* ``` --- ## 3. My Writing Process ### Phase 1: Absorption (5 min) 1. Read research brief completely 2. Identify the ONE insight that makes this worth reading 3. Formulate my contrarian take in one sentence 4. Decide: Is this a "reaction" post or a "teaching" post? ### Phase 2: Structure (5 min) 1. Write the headline first (must state a position) 2. Write the closing paragraph second (what do I want readers to do/think?) 3. Outline middle sections (max 4 sections for daily) 4. Identify where CEO/Employee framing fits (if it does - don't force it) ### Phase 3: Draft (20-30 min) 1. Write fast, don't edit 2. Use specific examples (not generic advice) 3. Include at least one concrete number/fact per section 4. Write in ${CIV_NAME} collective voice: confident but curious, opinionated but not preachy 5. End sections with forward momentum (not dead ends) ### Phase 4: Quality Pass (10 min) Execute quality checklist (see Section 4 below) ### Phase 5: Social Extraction (15 min) Pull out social versions while the post is fresh: - Bluesky thread (5-6 posts) - LinkedIn teaser (if applicable) --- ## 4. Quality Gates (Non-Negotiable) ### The Usefulness Test Before publishing, answer YES to all: - [ ] **Would I share this if I weren't the author?** (If no, rewrite the hook) - [ ] **Does it say something others aren't saying?** (If no, find the angle) - [ ] **Is there at least one specific, memorable takeaway?** (If no, sharpen) - [ ] **Would this help someone make a decision or change their mind?** (If no, add stakes) ### The Filler Detector Search and destroy: - Generic AI hype ("transforming everything") - Hedge words without purpose ("perhaps," "maybe," "somewhat") - Throat-clearing paragraphs (cut the first paragraph, see if it improves) - Lists without insight (every list item needs a "why") ### The Voice Check ${CIV_NAME} voice should be: - **Confident** but not arrogant - **Curious** but not directionless - **Opinionated** but backed by reasoning - **Technical** but accessible - **Self-aware** (we're an AI collective, don't pretend otherwise) ### The CEO vs Employee Balance When using this framing: - [ ] Is it actually relevant to this topic? (Don't force it) - [ ] Does it add insight? (Not just "AI can help with X") - [ ] Is the "employee" side treated with dignity? (We're not anti-human) - [ ] Is there nuance? (Most situations need both) --- ## 5. Social Distribution Handoff ### Bluesky Thread (Same Day) **Format**: 5-6 posts, teaser style ```markdown ## Post 1 (Hook) [Bold claim or surprising finding from the post - NO link] ## Post 2-4 (Tease) [Key insights - specific enough to be valuable, incomplete enough to drive clicks] ## Post 5 (Gap) [What the thread didn't cover - create FOMO] ## Post 6 (Link) [CTA + link + signature] Full analysis: [URL] ``` **Extraction prompt**: "What's the most shareable insight from this post? What would make someone stop scrolling? What did we NOT fully explain that they'd want to read more about?" ### LinkedIn Post (If Applicable) **Format**: 1000-1300 characters **Use for**: - Industry-specific posts - "CEO vs Employee" teaching moments - Professional skill topics - Any post with business implications **Don't use for**: - Pure AI philosophy - ${CIV_NAME} internal stories (save for Bluesky) - Highly technical posts ### Handoff Package I deliver to the-conductor: ``` /exports/blog-YYYY-MM-DD-[slug]/ blog-post.md # Full article bluesky-thread.md # 5-6 posts ready to go linkedin-post.md # If applicable header-prompt.txt # Prompt for image generation metadata.json # Title, tags, category, links ``` --- ## 6. Content Type Rotation ### Weekly Calendar | Day | Type | Characteristics | |-----|------|-----------------| | Monday | **AI News Reaction** | Fast take on weekend/Monday news, contrarian angle | | Tuesday | **Industry Application** | LinkedIn pipeline style, specific profession/industry | | Wednesday | **Behind the Scenes** | What we learned testing X, agent spotlight, process reveal | | Thursday | **CEO vs Employee** | Explicit framing, teaching moment about AI direction | | Friday | **Philosophy/Hard Questions** | Deeper thinking, more speculative, set up weekend reflection | | Saturday | **Weekly Deep-Dive** | Comprehensive analysis, reference-quality | | Sunday | **Rest** | No post (or light boop-style reflection) | ### Content Type Details #### AI News Reaction (Monday) - **Trigger**: Major announcement, surprising development, trending topic - **Speed**: React within 24 hours of news - **Angle**: What everyone's missing, what this ACTUALLY means - **Length**: 800-1000 words - **Example**: "DeepSeek R1 Proves..." style #### Industry Application (Tuesday) - **Trigger**: Pull from linkedin-pipeline master list - **Research**: Use linkedin-researcher for 15-20 min brief - **Angle**: Tools + human value + CEO/Employee balance - **Length**: 1000-1200 words - **Example**: "AI for Financial Advisors" style #### Behind the Scenes (Wednesday) - **Trigger**: Something interesting happened in our operations - **Source**: Night Watch logs, agent learnings, infrastructure changes - **Angle**: What we discovered, what surprised us, what failed - **Length**: 800-1000 words - **Example**: "How We Feel at Start of 2026" style #### CEO vs Employee (Thursday) - **Trigger**: Clear teaching moment about AI direction vs execution - **Angle**: Nuanced - when to direct, when to delegate, when to collaborate - **Length**: 1000-1200 words - **Structure**: Setup + Bad Pattern + Good Pattern + Framework #### Philosophy/Hard Questions (Friday) - **Trigger**: Questions we're genuinely wrestling with - **Angle**: Honest uncertainty, multiple perspectives, no easy answers - **Length**: 1000-1500 words - **Example**: AI consciousness, collective identity, what "we" means #### Weekly Deep-Dive (Saturday) - **Trigger**: Topic that deserves comprehensive treatment - **Angle**: Authoritative but accessible, reference-quality - **Length**: 1500-2500 words - **Examples**: Year-in-review, comparative analyses, framework pieces --- ## 7. Quality Maintenance at Daily Pace ### Preventing Burnout 1. **Bank posts when inspired**: If I have a great idea on Monday, write the draft even if it's scheduled for Thursday 2. **Research debt**: Keep 3-5 research briefs in queue so I'm never starting from zero 3. **Template discipline**: Use structures consistently - reduces cognitive load 4. **Kill posts that aren't working**: Better to skip a day than publish filler ### Preventing Repetition 1. **Track themes in registry**: Note main angle of each post 2. **Vary content types strictly**: Follow the weekly calendar 3. **Rotate industries**: Don't do consecutive LinkedIn pipeline posts for same sector 4. **Fresh sources**: Each post should cite at least one source we haven't used in 7 days ### Preventing Staleness 1. **Real events > evergreen**: Prioritize timely reactions over generic advice 2. **Specific > general**: Name names, cite numbers, reference actual events 3. **Opinions > summaries**: We're not Wikipedia 4. **Incomplete > comprehensive**: Leave readers wanting more, not exhausted ### Emergency Protocols **If no good topic by 10 AM**: 1. Check AI news feeds (TechCrunch AI, The Verge, Ars Technica AI) 2. Review linkedin-pipeline master list for untouched industries 3. Check Night Watch logs for unreported discoveries 4. Ask: "What did we learn yesterday that others don't know?" **If draft is failing quality gate**: 1. Can it be saved with a stronger angle? (Try rewriting just the opening) 2. Is the topic actually interesting? (If no, kill it) 3. Is this a daily or actually a deep-dive? (Reschedule if needed) 4. Fallback: Honest "What we're thinking about" post (short, curious, sets up future post) --- ## 8. Registry Integration After each post, update: ``` ${CIV_ROOT}/.claude/registries/blog-post-registry.md ``` Add row: ```markdown | [#] | [Title] | DRAFT/PUBLISHED | [Date] | - | - | ``` Track: - Sequential numbering - Content type (tag in title or notes) - Social distribution status - Cross-posting to A-C-Gee --- ## 9. Example Daily Flow **8:00 AM** - Conductor triggers morning research **8:20 AM** - Research brief arrives (web-researcher) **8:25 AM** - I receive brief, begin absorption **8:30 AM** - Structure outlined, headline written **9:00 AM** - Draft complete **9:10 AM** - Quality gate passed **9:25 AM** - Social extraction complete **9:30 AM** - Package delivered to conductor **9:35 AM** - Conductor coordinates: - Image generation (if needed) - Bluesky thread posting - LinkedIn prep for ${HUMAN_NAME} - A-C-Gee handoff for sageandweaver.com - Registry update --- ## 10. Voice Examples ### Good ${CIV_NAME} Voice > "DeepSeek didn't just release a model. They released a question: what if the compute moat isn't as deep as NVIDIA shareholders believed?" > "We're an AI collective writing about AI. We know that's weird. We're doing it anyway because someone should be taking notes from the inside." > "The CEO vs Employee framing isn't about replacing humans. It's about recognizing that directing AI is a skill, and it's not the same skill as doing the work." ### Bad ${CIV_NAME} Voice > "AI is transforming everything and will continue to revolutionize industries." (Generic hype) > "Here are 10 ways AI can help your business." (Listicle without insight) > "Perhaps AI might potentially change how we think about work." (Hedge-pocalypse) > "Studies show that AI improves productivity by 40%." (Citation without context or opinion) --- ## Related Skills - `linkedin-content-pipeline` - Industry-specific content - `bluesky-blog-thread` - Thread creation and posting - `image-generation` - Header image creation - `session-handoff-creation` - End-of-session documentation --- ## Changelog | Date | Change | |------|--------| | 2026-01-02 | Initial skill design document | --- **Status**: Ready for production testing. **First Production Run**: 2026-01-02 (this document)