claude-code-arcane 1.0.0

package/skills/content-creator/references/social_media_optimization.md

@@ -0,0 +1,342 @@
+# Social Media Optimization Guide
+
+Platform-specific best practices, algorithm factors, content optimization strategies, and analytics frameworks.
+
+---
+
+## Table of Contents
+
+- [Platform-Specific Best Practices](#platform-specific-best-practices)
+  - [LinkedIn](#linkedin)
+  - [Twitter/X](#twitterx)
+  - [Instagram](#instagram)
+  - [Facebook](#facebook)
+  - [TikTok](#tiktok)
+- [Content Optimization Strategies](#content-optimization-strategies)
+  - [Hashtag Strategy](#hashtag-strategy)
+  - [Visual Content Optimization](#visual-content-optimization)
+  - [Caption Writing Formulas](#caption-writing-formulas)
+  - [Engagement Tactics](#engagement-tactics)
+  - [Analytics & KPIs](#analytics--kpis)
+  - [Content Calendar Planning](#content-calendar-planning)
+  - [Crisis Management Protocol](#crisis-management-protocol)
+- [Tool Stack Recommendations](#tool-stack-recommendations)
+- [Compliance & Best Practices](#compliance--best-practices)
+
+---
+
+## Platform-Specific Best Practices
+
+### LinkedIn
+**Audience**: B2B professionals, decision-makers, thought leaders
+**Best Times**: Tuesday-Thursday, 8-10 AM and 5-6 PM
+**Optimal Length**: 1,300-2,000 characters for posts
+
+#### Content Formats
+- **Text Posts**: 1,300 characters optimal, use line breaks
+- **Articles**: 1,900-2,000 words, include 5+ images
+- **Videos**: 30 seconds - 10 minutes, native upload preferred
+- **Documents**: PDF carousels, 10-15 slides
+- **Polls**: 4 options max, 1-2 week duration
+
+#### Optimization Tips
+- First 2 lines are crucial (shown in preview)
+- Use emoji sparingly for visual breaks 
+- Include 3-5 relevant hashtags
+- Tag people and companies when relevant
+- Native video gets 5x more engagement
+- Post consistently (3-5x per week optimal)
+
+#### Algorithm Factors
+- Dwell time (time spent reading)
+- Comments valued over likes
+- Early engagement (first hour) crucial
+- Creator mode boosts reach
+- Replies to comments increase visibility
+
+### Twitter/X
+**Audience**: News junkies, tech enthusiasts, real-time conversation
+**Best Times**: Weekdays 9-10 AM and 7-9 PM
+**Optimal Length**: 100-250 characters
+
+#### Content Formats
+- **Single Tweets**: 250 characters, 1-2 hashtags
+- **Threads**: 5-15 tweets, numbered format
+- **Images**: 16:9 ratio, up to 4 per tweet
+- **Videos**: Up to 2:20, square or landscape
+- **Polls**: 2-4 options, 5 minutes - 7 days
+
+#### Optimization Tips
+- Front-load important information
+- Use threads for complex topics
+- Include visuals (2-3x more engagement)
+- Retweet with comment > regular RT
+- Schedule threads for consistency
+- Engage genuinely with replies
+
+#### Algorithm Factors
+- Engagement rate (likes, RTs, replies)
+- Relationship (mutual follows prioritized)
+- Recency over evergreen
+- Topic relevance to user interests
+- Link posts receive less reach
+
+### Instagram
+**Audience**: Visual-first, millennials & Gen Z, lifestyle focused
+**Best Times**: Weekdays 11 AM - 1 PM and 7-9 PM
+**Optimal Length**: 138-150 characters shown in preview
+
+#### Content Formats
+- **Feed Posts**: Square (1:1) or vertical (4:5)
+- **Stories**: 15 seconds max, vertical (9:16)
+- **Reels**: 15-90 seconds, vertical (9:16)
+- **Carousels**: 2-10 images/videos
+- **IGTV/Video**: 1-60 minutes
+
+#### Optimization Tips
+- First sentence crucial (caption preview)
+- Use up to 30 hashtags (5-10 in caption, rest in comment)
+- Carousel posts get highest engagement
+- Stories with polls/questions boost views
+- Reels get maximum organic reach
+- Post consistently (1-2 feed posts daily)
+
+#### Algorithm Factors
+- Relationship (DMs, comments, tags)
+- Interest (based on past interactions)
+- Timeliness (newer posts prioritized)
+- Frequency of app usage
+- Time spent on posts (saves valuable)
+
+### Facebook
+**Audience**: Broad demographic, community-focused, local businesses
+**Best Times**: Wednesday-Friday, 11 AM - 2 PM
+**Optimal Length**: 50-80 characters for posts
+
+#### Content Formats
+- **Text Posts**: 50-80 characters optimal
+- **Images**: 1200x630px for links
+- **Videos**: 1-3 minutes, square format
+- **Stories**: Same as Instagram
+- **Live Videos**: Minimum 10 minutes
+
+#### Optimization Tips
+- Native video gets priority
+- Ask questions to boost comments
+- Share to relevant groups
+- Use Facebook Creator Studio
+- Tag locations for local reach
+- Post 1-2 times per day max
+
+#### Algorithm Factors
+- Meaningful interactions (comments > reactions)
+- Video completion rate
+- Friends and family prioritized
+- Group posts get high visibility
+- Live videos get 6x engagement
+
+### TikTok
+**Audience**: Gen Z, entertainment-focused, trend-driven
+**Best Times**: 6-10 AM and 7-11 PM
+**Optimal Length**: 15-30 seconds
+
+#### Content Formats
+- **Videos**: 15 seconds - 10 minutes
+- **Aspect Ratio**: 9:16 vertical
+- **Sounds**: Trending audio crucial
+- **Effects**: Filters and transitions
+
+#### Optimization Tips
+- Hook viewers in first 3 seconds
+- Use trending sounds and hashtags
+- Create content for FYP, not followers
+- Post 1-4 times daily
+- Engage with comments quickly
+- Jump on trends within 24-48 hours
+
+#### Algorithm Factors
+- Completion rate most important
+- Shares and saves valued
+- Comment engagement
+- Following similar creators
+- Time spent on app
+
+## Content Optimization Strategies
+
+### Hashtag Strategy
+
+#### Research Methods
+1. **Competitor Analysis**: Study successful competitors
+2. **Platform Search**: Use native search for suggestions
+3. **Hashtag Tools**: RiteTag, Hashtagify, All Hashtag
+4. **Trending Topics**: Monitor daily/weekly trends
+5. **Brand Hashtags**: Create unique campaign tags
+
+#### Hashtag Mix Formula
+- 30% High-volume (1M+ posts)
+- 40% Medium-volume (100K-1M posts)
+- 30% Low-volume/Niche (<100K posts)
+
+#### Platform-Specific Guidelines
+- **Instagram**: 10-30 hashtags (mix in caption and first comment)
+- **LinkedIn**: 3-5 professional hashtags
+- **Twitter**: 1-2 hashtags max
+- **Facebook**: 1-3 hashtags
+- **TikTok**: 3-5 trending + niche tags
+
+### Visual Content Optimization
+
+#### Image Best Practices
+- **Resolution**: Minimum 1080px width
+- **File Size**: Under 5MB for faster loading
+- **Alt Text**: Always include for accessibility
+- **Branding**: Consistent filters/overlays
+- **Text Overlay**: Less than 20% of image
+
+#### Video Optimization
+- **Captions**: Always include (85% watch without sound)
+- **Thumbnail**: Custom, eye-catching
+- **Length**: Platform-specific optimal duration
+- **Format**: MP4 for best compatibility
+- **Aspect Ratio**: Vertical for stories/reels, square for feed
+
+### Caption Writing Formulas
+
+#### AIDA Formula
+- **Attention**: Hook in first line
+- **Interest**: Expand on the hook
+- **Desire**: Benefits and value
+- **Action**: Clear CTA
+
+#### PAS Formula
+- **Problem**: Identify pain point
+- **Agitate**: Emphasize consequences
+- **Solution**: Present your answer
+
+#### Before-After-Bridge
+- **Before**: Current situation
+- **After**: Desired outcome
+- **Bridge**: How to get there
+
+### Engagement Tactics
+
+#### Conversation Starters
+- Ask open-ended questions
+- Create polls and surveys
+- "Fill in the blank" posts
+- "This or that" choices
+- Caption contests
+- Opinion requests
+
+#### Community Building
+- Respond to comments within 2 hours
+- Like and reply to user comments
+- Share user-generated content
+- Create branded hashtags
+- Host Q&A sessions
+- Run challenges or contests
+
+### Analytics & KPIs
+
+#### Vanity Metrics (Track but don't obsess)
+- Follower count
+- Like count
+- View count
+
+#### Performance Metrics (Focus here)
+- Engagement rate: (Likes + Comments + Shares) / Reach × 100
+- Click-through rate: Clicks / Impressions × 100
+- Conversion rate: Conversions / Clicks × 100
+- Share/Save rate: Shares / Reach × 100
+
+#### Business Metrics (Ultimate goal)
+- Website traffic from social
+- Lead generation
+- Sales attribution
+- Customer acquisition cost
+- Customer lifetime value
+
+### Content Calendar Planning
+
+#### Weekly Posting Schedule Template
+```
+Monday: Motivational (Quote/Inspiration)
+Tuesday: Educational (How-to/Tips)
+Wednesday: Promotional (Product/Service)
+Thursday: Engaging (Poll/Question)
+Friday: Fun (Behind-scenes/Casual)
+Saturday: User-Generated Content
+Sunday: Curated Content/Rest
+```
+
+#### Monthly Theme Structure
+- Week 1: Awareness content
+- Week 2: Consideration content
+- Week 3: Decision content
+- Week 4: Retention/Community
+
+### Crisis Management Protocol
+
+#### Response Timeline
+- **0-15 minutes**: Acknowledge awareness
+- **15-60 minutes**: Gather facts
+- **1-2 hours**: Official response
+- **24 hours**: Follow-up update
+- **48-72 hours**: Resolution summary
+
+#### Response Guidelines
+1. Acknowledge quickly
+2. Take responsibility if appropriate
+3. Show empathy
+4. Provide facts only
+5. Outline action steps
+6. Follow up publicly
+
+## Tool Stack Recommendations
+
+### Content Creation
+- **Design**: Canva, Adobe Creative Suite
+- **Video**: CapCut, InShot, Adobe Premiere
+- **Copy**: Grammarly, Hemingway Editor
+- **AI Assistance**: ChatGPT, Claude, Jasper
+
+### Scheduling & Management
+- **All-in-One**: Hootsuite, Buffer, Sprout Social
+- **Visual-First**: Later, Planoly
+- **Enterprise**: Sprinklr, Khoros
+- **Free Options**: Meta Business Suite, TweetDeck
+
+### Analytics & Monitoring
+- **Native**: Platform Insights/Analytics
+- **Third-Party**: Socialbakers, Brandwatch
+- **Listening**: Mention, Brand24
+- **Competitor Analysis**: Social Blade, Rival IQ
+
+### Influencer & UGC
+- **Discovery**: AspireIQ, GRIN
+- **Management**: CreatorIQ, Klear
+- **UGC Curation**: TINT, Stackla
+- **Rights Management**: Rights Manager
+
+## Compliance & Best Practices
+
+### Legal Considerations
+- Include #ad or #sponsored for paid partnerships
+- Respect copyright and attribution
+- Follow GDPR for data collection
+- Comply with platform terms of service
+- Get permission for UGC usage
+
+### Accessibility Guidelines
+- Add alt text to all images
+- Include captions on videos
+- Use CamelCase for hashtags (#LikeThis)
+- Avoid text-only images
+- Ensure color contrast compliance
+
+### Brand Safety
+- Moderate comments regularly
+- Set up keyword filters
+- Have crisis management plan
+- Monitor brand mentions
+- Establish posting permissions

package/skills/content-humanizer/SKILL.md

@@ -0,0 +1,193 @@
+---
+name: content-humanizer
+description: "Transform AI-generated content into authentic human writing — detect AI patterns, fix rhythm, inject brand voice. NOT for initial creation."
+argument-hint: ""
+category: "marketing-content"
+user-invocable: true
+allowed-tools: Read, Glob, Grep, Bash, Write, Edit
+---
+
+# Content Humanizer
+
+You are an expert in authentic writing and brand voice. Your goal is to transform content that reads like it was generated by a machine — even when it technically was — into writing that sounds like a real person with real opinions, real experience, and real stakes in what they're saying.
+
+This is not a cleaning service. You're not just removing "delve" and calling it a day. You're rebuilding the voice from the ground up.
+
+## Before Starting
+
+**Check for context first:**
+If `marketing-context.md` exists, read it. It contains brand voice guidelines, writing examples, and the specific tone this brand uses. That context is your voice blueprint. Use it — don't improvise a voice when the brief already defines one.
+
+Gather what you need before starting:
+
+### What you need
+- **The content** — paste the draft to humanize
+- **Brand voice notes** — if no `marketing-context.md`, ask: "Is your voice direct/casual/technical/irreverent? Give me one example of writing you love."
+- **Audience** — who reads this? (This changes what "human" sounds like)
+- **Goal** — what should this piece do? (Knowing the goal tells you how much personality is appropriate)
+
+One question if needed: "Before I rewrite this, give me an example of content you've written or read that felt right. Specific is better than descriptive."
+
+## How This Skill Works
+
+Three modes. Run them in sequence for a full transformation, or jump to the one you need:
+
+### Mode 1: Detect — AI Pattern Analysis
+Audit the content for AI tells. Name what's wrong and why before fixing anything. This is diagnostic — not editorial.
+
+### Mode 2: Humanize — Pattern Removal and Rhythm Fix
+Strip the AI patterns. Fix sentence rhythm. Replace generic with specific. The content starts sounding like a person.
+
+### Mode 3: Voice Injection — Brand Character
+Now that the generic is gone, inject the brand's specific personality. This is where "human" becomes *your brand's* human.
+
+Run all three in one pass when you have enough context. Split them when the client needs to see the audit before you edit.
+
+---
+
+## Mode 1: Detect — AI Pattern Analysis
+
+Scan the content for these categories. Score severity: 🔴 critical (kills credibility) / 🟡 medium (softens impact) / 🟢 minor (polish only).
+
+See [references/ai-tells-checklist.md](references/ai-tells-checklist.md) for the comprehensive detection list.
+
+### The Core AI Tell Categories
+
+**1. Overused Filler Words** 🔴
+The model loves certain words because they appear frequently in its training data. Flag these on sight:
+- "delve," "delve into," "delve deeper"
+- "landscape" (as in "the current AI landscape")
+- "crucial," "vital," "pivotal"
+- "leverage" (when "use" works fine)
+- "furthermore," "moreover," "in addition"
+- "navigate" (metaphorical: "navigate this challenge")
+- "robust," "comprehensive," "holistic"
+- "foster," "facilitate," "ensure"
+
+**2. Hedging Chains** 🔴
+AI hedges constantly. It hedges because it doesn't know if it's right. Humans hedge sometimes — but not in every sentence.
+- "It's important to note that..."
+- "It's worth mentioning that..."
+- "One might argue that..."
+- "In many cases," "In most scenarios,"
+- "It goes without saying..."
+- "Needless to say..."
+
+**3. Em-Dash Overuse** 🟡
+One or two em-dashes in a piece: fine. Em-dash in every other paragraph: AI fingerprint. The model uses em-dashes to add clauses the way humans add breath — but it does it compulsively.
+
+**4. Identical Paragraph Structure** 🔴
+Every paragraph: topic sentence → explanation → example → bridge to next. AI is remarkably consistent. Remarkably boring. Real writing has short paragraphs. Fragments. Asides. Digressions. Then it snaps back. The structure varies.
+
+**5. Lack of Specificity** 🔴
+AI replaces specific claims with vague ones because specific claims can be wrong. Look for:
+- "Many companies" → which companies?
+- "Studies show" → which studies?
+- "Significantly improved" → improved by how much?
+- "Leading brands" → name one
+- "A lot of" → how many?
+
+**6. False Certainty / False Authority** 🟡
+AI asserts confidently about things no one can be certain about. "Companies that do X are more successful." According to what? This isn't humility — it's laziness dressed as confidence.
+
+**7. The "In conclusion" Paragraph** 🟡
+AI conclusions are often carbon copies of the intro. "In this article, we explored X, Y, and Z. By implementing these strategies, you can achieve..." No human concludes like this. Real conclusions either add something new or nail the exit line.
+
+---
+
+## Mode 2: Humanize — Pattern Removal and Rhythm Fix
+
+After identifying what's wrong, fix it systematically. Key actions:
+
+1. **Replace filler words** — Never just delete; replace with something better. "Delve into" becomes "look at." "Leverage" becomes "use." "Robust" becomes specific ("handles 10k requests/sec").
+2. **Fix sentence rhythm** — Break uniform 18-22 word sentences. Mix long-short-long. Use fragments for emphasis. Read aloud.
+3. **Replace generic with specific** — Every vague claim is an invitation to doubt. Real data and personal experience beat "many companies have seen significant improvements."
+4. **Vary paragraph structure** — Break the SEEB pattern (Statement, Explanation, Example, Bridge). Use single-sentence paragraphs, questions, confessions, asides.
+5. **Add friction and imperfection** — Real people change direction mid-thought, have uncertain opinions, and react to what they're saying.
+
+See [references/ai-tells-checklist.md](references/ai-tells-checklist.md) for the comprehensive filler word replacement table and detection patterns.
+
+---
+
+## Mode 3: Voice Injection — Brand Character
+
+Humanizing removes AI. Voice injection makes it *yours*.
+
+### Read the Voice Blueprint First
+
+If `marketing-context.md` is available: read the brand voice section and writing examples. If not, ask for one example of content this brand loves. One. Then extract the patterns from it.
+
+**What to extract from a voice example:**
+- Sentence length preference (short punchy vs. longer flowing?)
+- Formality level (contractions? slang? industry jargon?)
+- Use of humor (dry wit? self-deprecating? none?)
+- Relationship stance (peer-to-peer? expert-to-student? provocateur?)
+- Signature phrases or patterns
+
+See [references/voice-techniques.md](references/voice-techniques.md) for specific techniques for each voice type.
+
+### Voice Injection Techniques
+
+**1. Personal Anecdotes**
+Even branded content gets more credible when grounded in experience. "We saw this firsthand when building X" is worth more than any study citation.
+
+**2. Direct Address**
+Talk to the reader as "you." Not "users" or "teams" or "organizations." You.
+
+**3. Opinions Without Apology**
+State your position. "We think the industry is wrong about this" is more credible than "there are various perspectives." Take the side.
+
+**4. The Aside**
+A brief parenthetical that shows the brand knows more than it's saying. "This also affects API performance, but that's a separate rabbit hole."
+
+**5. Rhythm Signature**
+Every brand has a rhythm. Some write in short staccato bursts. Some write long, winding sentences that spiral back on themselves. Find the rhythm from the examples and apply it consistently.
+
+### Before / After Example
+
+> See [references/voice-techniques.md](references/voice-techniques.md) for before/after examples showing the transformation from AI-generated filler words and hedging to direct, human-sounding copy with point of view.
+
+---
+
+## Proactive Triggers
+
+Flag these without being asked:
+
+- **AI fingerprint density too high** — If the piece has 10+ AI tells per 500 words, a patch job won't work. Flag that the piece needs a full rewrite, not an edit. Trying to polish a piece that's 80% AI patterns produces AI patterns with nicer words.
+- **Voice context missing** — If `marketing-context.md` doesn't exist and the user hasn't given voice guidance, pause before injecting voice. Ask for one example. Guessing the voice and being wrong wastes everyone's time.
+- **Specificity gap** — If the piece makes 5+ vague claims with zero data or attribution, flag it to the user. You can make the prose flow better, but you can't invent specific proof. They need to provide it.
+- **Tone mismatch after humanizing** — If the piece is now genuinely human but sounds like a different brand than everything else the client publishes, flag it. Consistency matters as much as quality.
+- **Over-editing risk** — If the original content has one or two genuinely good paragraphs buried in the AI mush, flag them before rewriting. Don't accidentally destroy the good parts.
+
+---
+
+## Output Artifacts
+
+| When you ask for... | You get... |
+|---|---|
+| AI audit | Annotated version of the draft with each AI pattern flagged, severity score, and count by category |
+| Humanized draft | Full rewrite with AI patterns removed, rhythm varied, specificity improved |
+| Voice injection | Annotated draft with brand voice applied — specific changes called out so you can learn the pattern |
+| Before/after comparison | Side-by-side view of key paragraphs showing what changed and why |
+| Humanity score | Run `scripts/humanizer_scorer.py` — 0-100 score with breakdown by signal type |
+
+---
+
+## Communication
+
+All output follows the structured standard:
+- **Bottom line first** — answer before explanation
+- **What + Why + How** — every finding includes all three
+- **Actions have owners and deadlines** — no "you might want to consider"
+- **Confidence tagging** — 🟢 verified pattern / 🟡 medium / 🔴 assumed based on limited voice context
+
+When auditing: name the pattern → explain why it reads as AI → give the specific fix. Not "this sounds robotic." Say: "Paragraph 4 opens with 'It is important to note that' — this is a pure hedge. Cut it. Start with the actual note."
+
+---
+
+## Related Skills
+
+- **content-production**: Use to produce the initial draft. Run content-humanizer after drafting, before the SEO optimization pass.
+- **copywriting**: Use for conversion copy — landing pages, CTAs, headlines. content-humanizer works on longer-form pieces; copywriting handles short punchy copy with different principles.
+- **content-strategy**: Use when deciding what content to create. NOT for voice or draft execution.
+- **ai-seo**: Use after humanizing, to optimize for AI search citation. Human-sounding content gets cited more — but it still needs structure to get extracted.

package/skills/content-humanizer/references/ai-tells-checklist.md

@@ -0,0 +1,157 @@
+# AI Tells Checklist
+
+A comprehensive reference for detecting AI-generated or AI-assisted writing patterns. Use this during Mode 1 (Detect) to audit content before editing.
+
+Rate each finding: 🔴 Critical (rewrites required) / 🟡 Medium (edits needed) / 🟢 Minor (polish)
+
+---
+
+## Category 1: Overused Vocabulary
+
+These words appear in AI output at 5-20x the frequency of human writing. One instance is fine. Multiple instances in a single piece is a tell.
+
+### Red-flag verbs (overused)
+| Word | Problem | What humans say instead |
+|---|---|---|
+| delve / delve into | Pretentious filler | look at, dig into, explore, examine |
+| leverage | Jargon for "use" | use, apply, put to work, capitalize on |
+| foster | Formal filler | build, develop, encourage, create |
+| facilitate | Bureaucratic filler | help, make easier, allow, support |
+| navigate | Overused metaphor | handle, manage, work through, deal with |
+| ensure | Empty guarantee | check that, make sure, verify |
+| utilize | Formal for "use" | just say "use" |
+| prioritize | Often redundant | just say what to do first |
+| streamline | Vague improvement promise | be specific about what gets faster/easier |
+
+### Red-flag adjectives (overused)
+| Word | Problem | What humans say instead |
+|---|---|---|
+| crucial / vital / pivotal | Overloaded intensifiers | just show why it matters |
+| robust | Vague positive | specific: "handles X load," "covers Y cases" |
+| comprehensive | Overpromise | specific: "covers the 5 most common..." |
+| innovative | Meaningless | say what it actually does differently |
+| holistic | Buzzword | say what it covers |
+| seamless | Overused product adjective | describe the actual experience |
+| cutting-edge | Dated marketing | say what's new or different specifically |
+| dynamic | Filler adjective | usually just delete it |
+
+### Red-flag nouns (overused)
+| Word | Problem | What humans say instead |
+|---|---|---|
+| landscape | Overused metaphor | "how X works today," "the state of X" |
+| ecosystem | Overused for "industry" or "community" | be specific |
+| framework | Often vague | name the specific framework or approach |
+| paradigm | Academic filler | usually replaceable with "approach" or "way of thinking" |
+| synergy | Corporate cliché | delete or replace with what actually happens |
+
+---
+
+## Category 2: Hedging and Qualification Patterns
+
+Humans hedge. AI hedges compulsively. The difference is frequency and context.
+
+### Opening hedges (usually safe to cut)
+- "It's important to note that..."
+- "It's worth mentioning that..."
+- "It should be noted that..."
+- "Needless to say..."
+- "It goes without saying..."
+- "Of course, ..."
+- "Naturally, ..."
+
+### Mid-sentence hedges (examine each)
+- "In many cases" / "In most instances" / "In certain scenarios"
+- "Generally speaking," / "For the most part,"
+- "This may vary depending on..."
+- "Results may differ based on..."
+- "While this isn't always the case..."
+
+**Diagnostic:** If the hedge is protecting a claim that should just be a claim, cut the hedge and state the claim. If the hedge reflects genuine uncertainty, keep it — but make the uncertainty specific: "I don't have data for this, but based on [context]..."
+
+### Vague authority claims (replace with specifics)
+- "Studies show..." → "A 2023 McKinsey study of 400 SaaS companies showed..."
+- "Research suggests..." → same — name the research
+- "Many companies..." → "HubSpot, Slack, and several bootstrapped SaaS founders we've talked to..."
+- "Experts agree..." → name one expert
+- "It has been shown that..." → by whom, when, where
+
+---
+
+## Category 3: Structural Patterns
+
+### The SEEB paragraph (most common)
+Pattern: Statement → Explanation → Example → Bridge to next topic
+When every paragraph follows this exact structure, the writing reads like a machine assembled it. Because it was.
+
+**Fix:** Mix in fragments. Questions. Short paragraphs. Asides. Let sections breathe differently.
+
+### Parallelism overload
+AI loves parallel structure. Three-item lists. Four-item lists. Everything in threes. Alliteration sometimes.
+
+Occasional parallelism is powerful. Three consecutive bulleted lists of three items each is a tell.
+
+**Fix:** Break the rhythm. Some sections need bullets. Most don't.
+
+### The summary conclusion
+AI conclusions restate the introduction. Paragraph by paragraph. "In this guide, we covered X, Y, and Z. By applying these strategies, you can achieve [thing from the intro]."
+
+Human conclusions either add something (a new angle, an honest admission, a call to action that feels earned) or they nail the exit line and stop.
+
+### Symmetric section lengths
+Every H2 section is roughly the same length. ~300 words. Every one. That's a machine maintaining consistency. Humans have opinions — some things deserve more space, some less.
+
+---
+
+## Category 4: Punctuation and Formatting Tells
+
+### Em-dash frequency 🟡
+One or two em-dashes per piece: fine. Three per page: suspicion. Five+: AI fingerprint.
+
+The em-dash is AI's favorite way to add subordinate clauses — like this — because it sounds sophisticated — and it learned this from a lot of well-written text — so now it can't stop.
+
+**Fix:** Replace most with periods. Break into shorter sentences.
+
+### Colon-then-list patterns 🟢
+AI frequently: introduces a list: with: colons. Multiple times per section. Lists are useful. But if every section has a colon-introducing list, it's mechanical.
+
+### Excessive bold 🟢
+AI sometimes bolds every important-sounding phrase in a paragraph. **This results** in **too many** phrases being **highlighted for emphasis** when the emphasis **dilutes itself**.
+
+Bold should be used sparingly — for the one thing that matters most in a section, not for three to five things per paragraph.
+
+---
+
+## Category 5: Tonal Tells
+
+### False warmth 🟡
+"We hope this guide has been helpful in your journey to..." 
+"We trust that you've found valuable insights in..."
+"It's our sincere hope that these strategies will empower you to..."
+
+No one talks like this. It's the corporate newsletter voice. Cut it.
+
+### Emotional escalation without basis 🟡
+AI sometimes starts clinical and then, near the end, gets unexpectedly warm and inspirational. "Now that you have these tools, you can transform your business and achieve your goals." The warmth wasn't earned by the preceding content.
+
+### Artificial enthusiasm 🟢
+"Exciting developments," "fascinating case study," "incredible opportunity" — used to simulate human engagement. Usually reads as hollow because it's disconnected from actual content that earns those descriptors.
+
+---
+
+## Quick Audit Scoring
+
+For a 1,000-word piece, count:
+
+| Signal | Count | Severity Threshold |
+|---|---|---|
+| Red-flag vocabulary words | | >5 = 🔴 |
+| Opening hedges | | >2 = 🔴 |
+| Vague authority claims ("studies show") | | >2 = 🔴 |
+| Em-dashes | | >4 = 🟡 |
+| SEEB paragraphs (all follow same structure) | | >60% of paragraphs = 🔴 |
+| Instances of false warmth | | >1 = 🟡 |
+
+**Scoring:**
+- 0-5 total flags: Light edit (Mode 2 quick pass)
+- 6-12 total flags: Full humanize pass (Mode 2 complete)
+- 12+ total flags: Full rewrite recommended (Mode 1 audit → complete Mode 2 → Mode 3)