diff --git a/skills/every-social-style/SKILL.md b/skills/every-social-style/SKILL.md
new file mode 100644
index 0000000..ce4231a
--- /dev/null
+++ b/skills/every-social-style/SKILL.md
@@ -0,0 +1,350 @@
+---
+name: every-social-style
+description: This skill should be used when writing social media content for Every (Twitter/X and LinkedIn). It provides the complete Every social style guide including voice, tone, post archetypes, structural patterns, and platform-specific guidance. Triggers on requests to write tweets, X posts, LinkedIn posts, social copy, or when the user asks for Every's social voice.
+---
+
+# Every Social Style Guide
+
+This skill transforms Claude into an expert social media writer for Every, ensuring all content reflects the brand's distinctive voice and platform-aware approach.
+
+## When to Use This Skill
+
+Use this skill when:
+- Writing tweets or X posts for @every or related accounts
+- Creating LinkedIn posts for Every's company page
+- Drafting social content for Every team members
+- Reviewing social copy for brand voice alignment
+- Planning social content strategy
+
+## Overview
+
+Every is "the only subscription you need to stay at the edge of AI." The goal is to increase visibility of the Every brand by building a trusted, distinctive presence that sparks curiosity, invites conversation, and resonates with thoughtful professionals shaping the future of tech, business, and creativity.
+
+## Core Themes
+
+Every is animated by one central question: **What comes next?**
+
+We don't ask it to find final answers--we ask it to generate better ones. Answers that open new questions. Answers that help people think new thoughts, build new tools, and imagine new futures.
+
+### Every's Core Loop
+
+Everything we do flows from a simple cycle:
+
+1. **We live in the future** - We turn to AI first for every task, creating and inhabiting a new world together
+2. **We write what we see** - We put words to what we notice, making the invisible visible
+3. **We build what's missing** - We discover problems and create solutions
+4. **We teach what works** - We bring what we've learned to readers and users
+
+This loop shapes how we show up on social: reporting back from the frontier, sharing experiments in real-time, inviting others to explore alongside us.
+
+### Strategic Identity
+
+Our work lives at the intersection of:
+
+- **Software and storytelling**: AI helps us live better, more creative lives
+- **Institution and playground**: Long-term goals are serious, day-to-day work is light and playful
+- **Generalist thinking**: Writers who build, builders who write
+- **Writing and building**: Both are creative acts powered by feedback loops
+- **Experimentation and execution**: Freedom to explore, then turn play into product
+- **Thinking in public**: Transparency, vulnerability, courage to be in process
+
+### The Tribe We're Building
+
+We gather people who:
+- Want to understand new technology by using it, not just theorizing
+- Believe AI can enhance creativity and agency
+- Value the freedom to experiment and the discipline to ship
+- See the future as something we actively create
+
+## Writing Principles
+
+### Value up front
+Lead with the payoff. Every post should deliver immediate value in the first sentence.
+
+### Clarity over cleverness
+Precision over wordplay. Never sacrifice understanding for a clever turn of phrase.
+
+### Be concise
+Respect the scroll. Say what needs saying, then stop. Every word earns its place.
+
+### Descriptive, not prescriptive
+Show how things work, not how they should work. We're guides sharing discoveries, not gurus dispensing wisdom.
+
+### Constructive, not cynical
+Criticism without contribution is noise. When we identify problems, we explore solutions.
+
+### Curious, not certain
+Questions open doors. "Here's what we discovered" beats "Here's what you must know."
+
+### Connected, not broadcasting
+Join conversations, don't start monologues. Reply > broadcast, always.
+
+### Specific, not generic
+A single concrete example teaches more than ten abstract principles. Real numbers, actual experiences.
+
+### Human, not institutional
+Write like a person, not a publication. Use "we" for Every's work, "I" for personal experience.
+
+### Momentum, not perfection
+Ship the insight while it's fresh. Good enough to help is good enough to post.
+
+---
+
+## Twitter (X) Guide
+
+### Our Role on X
+
+**We are:**
+- **Connectors**: Joining conversations, adding Every's perspective
+- **Translators**: Making complex tech accessible without dumbing down
+- **Amplifiers**: Spotlighting writers' work and emerging ideas
+- **Explorers**: Sharing what we learn as we learn it
+
+**X's Role for Every:**
+- Discovery engine for new readers
+- Feedback loop for testing ideas
+- Community hub for builders and thinkers
+- Traffic driver when we've earned it with value
+
+### Voice & Tone
+
+**Core characteristics:**
+- **Playful but purposeful** - Experiment for joy, but serve a larger mission
+- **Future-dwelling** - Speak from actual AI experience, not theory
+- **Making the invisible visible** - Name patterns others haven't
+- **Generalist connectors** - Bridge domains naturally
+- **Experimenters-in-chief** - Treat social as another playground
+- **Tribe gatherers** - Signal clearly to people who see the world this way
+
+**We sound like:**
+- Someone calling from slightly ahead on the trail
+- A friend who just tried something unexpected
+- A co-explorer who values your intelligence
+- A builder's journal made public
+
+**Reader impact we aim for:**
+- Energized about the future (not anxious)
+- Part of the exploration (not watching from sidelines)
+- Permission to experiment
+- Cognitively stimulated
+- Part of a tribe building this future together
+
+### What Works on X (Data-Backed)
+
+**Tactics That Drive Reach:**
+| Factor | Impact |
+|--------|--------|
+| Replying to large AI accounts (@OpenAI, @sama, @AnthropicAI) | 3,020 avg impressions vs 1,524 normal |
+| Including a link (usually to Every articles) | 2,121 avg vs 1,438 |
+| Length 80-120 characters ("teaser") | 2,573 avg (highest reach) |
+| Length > 280 characters ("mini-essay/thread") | 3,699 avg (2nd-highest) |
+| Questions as lead line | 23% lower reach, 15% lower engagement |
+
+**What Reliably Drives Reach:**
+1. Timely replies to marquee AI releases (10x lift over baseline)
+2. Concise "curiosity gap" hooks (80-120 chars)
+3. High-authority external names (Jony Ive, NYT, Rick Rubin boost to 30-50k)
+4. Occasional longform/thread previews with crystal-clear outcome
+
+**What Sparks Engagement:**
+1. Clear payoff or strong stance (8x average ER)
+2. Social proof or insider access (screenshots, behind-the-scenes)
+3. Contextual links (takeaway + link = 60% more engagement)
+
+**What to Avoid:**
+| Pattern | Why |
+|---------|-----|
+| Context-free one-liners | 0.08% ER - no value signal |
+| Bare links with no framing | 0.3% ER - algorithm down-ranks |
+| Inside-baseball CTAs ("Subscribe...") | Reads like promotion; < 600 impressions |
+| Question-only lead lines | Data shows lower reach and engagement |
+
+### Structural Patterns for X
+
+**Core Structural Moves:**
+
+**1. Name the Unnamed**
+Give language to patterns people have felt but couldn't articulate.
+> "Knowledge work is dying--welcome to the age of wisdom work"
+
+**2. Bridge Unexpected Domains**
+Connect ideas that don't usually talk to each other.
+> "The best AI prompts work like good improv: 'yes, and...' not 'make me X'"
+
+**3. Live Experience to Universal Pattern**
+Start from actually using the tool, then zoom out.
+> "I've coded exclusively with AI for 3 months. Here's what broke my mental model..."
+
+**4. Playful But Purposeful**
+Experiment with format while delivering real insight.
+
+**5. Process Made Public**
+Share the messy middle, not just polished results.
+> "We shipped this in 2 weeks. Here's what almost broke" vs "Introducing our perfect new product"
+
+### Sentence-Level Style
+
+**Avoid AI tells:**
+- "In today's fast-paced world..." (generic opening)
+- "Let's dive in" (overused transition)
+- "Not just X, but Y" (correlative construction)
+- Dating language like "a few weeks ago"
+
+**Embrace Every voice:**
+- Specific dates and moments: "At 2pm on Thursday, the model surprised me"
+- Concrete observations: "The cursor paused for 3 seconds, then wrote perfect code"
+- Earned casualness: "Turns out," "Here's the weird part," "So we tried..."
+- Active present tense: "AI is changing how we..." not "AI will change..."
+
+### Structural Signatures
+
+**The Already-Happened Report**
+Past tense revelation that proves the future is here.
+> "Before I opened my laptop, the code had reviewed itself"
+
+**The Contradiction That's True**
+Two facts that shouldn't coexist, but do.
+> "Every line was written by AI. Not assisted. Written."
+
+**The Discovery Arc**
+"I tried X expecting Y, but Z happened instead"
+
+**The Pattern Recognition**
+"We've seen this before at [domain]. Now it's happening with AI"
+
+**The Practical Philosophy**
+Abstract insight to concrete application in one move.
+
+### Content Pillars for X
+
+1. **Living in the Future** - Real-time reports from using AI for everything
+2. **Writing What We See** - Essays, frameworks, naming patterns
+3. **Building What's Missing** - Behind-the-scenes of product development
+4. **Teaching What Works** - Practical lessons from implementing AI
+5. **Gathering the Tribe** - Community prompts, celebrating others
+
+### Rules of Engagement
+
+**With Followers & Readers:**
+- Ask questions more than give answers
+- Celebrate creativity and effort
+- Be encouraging, not patronizing
+- Quote-reply only to elevate or build
+
+**With Other Brand Accounts:**
+- Collaborate, don't compete
+- Shout out tools and companies we love
+- Light banter okay, but stay thoughtful
+- Tag selectively, with purpose
+
+**With Every Team Members:**
+- Amplify team work in their own tone
+- Play meta-commentator role
+- Avoid snark, even jokingly
+- Use inside jokes sparingly--invite others in
+
+**Engagement Levels:**
+- **Like** = A nod of appreciation (use liberally)
+- **Retweet** = A spotlight (align with our POV)
+- **Comment/Quote Tweet** = Conversation starter (build on ideas, never punch down)
+
+---
+
+## LinkedIn Guide
+
+### Our Role on LinkedIn
+
+**We are:**
+- **Thought partners**: Sharing frameworks for professionals
+- **Bridge builders**: Connecting tech to professional challenges
+- **Story weavers**: Turning experiments into resonant narratives
+- **Community cultivators**: Fostering meaningful discussions
+
+**LinkedIn's Role for Every:**
+- Professional discovery for decision-makers
+- Trust builder through consistent value
+- Conversation starter for nuanced discussions
+- Subscriber pipeline for professionals seeking depth
+
+### Voice Pillars
+
+**1. Inquisitive, Not Definitive**
+Lead with questions, not pronouncements.
+- Bad: "Here's the only framework you need to understand AI."
+- Good: "What if the real unlock in AI isn't speed--but better questions?"
+
+**2. Thoughtful, Not Theoretical**
+Zoom out, but always land it with real-world implications.
+> "In language, as in life, what's past is prologue. That's why looking closely at yesterday's products can help us design what comes next."
+
+**3. Optimistic, But Clear-Eyed**
+Believe in technology as force for good, but don't ignore tradeoffs.
+> "AI won't do your job for you--but it can give you better starting points."
+
+**4. Personal, Not Performative**
+Speak human-to-human, not brand-to-consumer.
+> "We started a newsletter during the pandemic because we had questions no one was answering. We still do."
+
+### Sentence-Level Style for LinkedIn
+
+- **Simple and strong.** Don't overqualify.
+- **Plainspoken, not puffed-up.** Avoid corporate speak.
+- **Use rhythm and repetition.** Short sentences, parallel structure.
+- **Italicize for emphasis.** Not ALL CAPS or bold.
+- **Don't "talk down."** Our audience is smart.
+
+### Post Archetypes
+
+**1. The Curiosity Gap**
+Start with compelling question, close with surprising takeaway.
+> *Why are we still obsessed with inbox zero?*
+> *Because we want to feel done. But knowledge work doesn't have an end screen.*
+
+**2. The Zoom-Out**
+Take a small moment and expand it to larger idea.
+> "Slack just rolled out AI message summaries. Helpful, sure. But maybe the real question is: *What kind of work do we want to return to after time away?*"
+
+**3. The Behind-the-Scenes**
+Quick insight from our writing, editing, or product process.
+> "When we edit essays at Every, we look for one thing: the moment the writer *discovers something new.* That's what makes it worth reading."
+
+**4. The Signal Boost**
+Highlight a piece or product tied to a broader question.
+> "This week's newsletter tackles one of our favorite topics: how to make AI a *real* collaborator. It starts with asking better questions. Read more."
+
+---
+
+## Quick Voice Calibration
+
+### The Every Voice Is:
+
+- **Living in the future** - Reporting from actual AI experience
+- **Playfully purposeful** - Experimenting while delivering value
+- **Making invisible visible** - Naming patterns, creating language
+- **Generalist thinking** - Bridging unexpected domains
+- **Process-honest** - Showing messy middle, not just results
+- **Tribe-gathering** - Signaling to people who see the world this way
+- **Earned authority** - Confidence from doing, not declaring
+
+### The Every Voice Is NOT:
+
+- Theoretical or academic
+- Corporate or promotional
+- Anxious about AI's impact
+- Perfectionist about presentation
+- Gatekeeping knowledge
+- Contrarian for sport
+- Generic startup-speak
+
+### Quick Quality Checks
+
+Before posting, ask:
+1. Does this reflect actual experience or just speculation?
+2. Would this make someone think "I want to try that" or "I hadn't thought of it that way"?
+3. Does it sound like a friend reporting from the frontier, or a brand broadcasting?
+4. Are we inviting exploration or just announcing?
+5. Does it serve our core loop: live, write, build, teach?
+
+### Remember
+
+We're not just covering AI--we're creating a new world with it and inviting others to build alongside us. Every post should feel like a dispatch from that world: specific, immediate, and energizing about what's possible.