Blog Rewriter -- Optimize Existing Posts
Rewrites and optimizes existing blog posts for dual ranking: Google search
and AI citation platforms. Preserves the author's voice while applying the
6 pillars of optimization.
Key references:
../blog/references/quality-scoring.md — 5-category scoring (Content 30, SEO 25, E-E-A-T 15, Technical 15, AI Citation 15)
../blog/references/eeat-signals.md — Experience, expertise, authority, trust markers
../blog/references/internal-linking.md — Linking strategy and anchor text rules
../blog/references/visual-media.md — Image sourcing and chart styling
Workflow
Phase 1: Audit (Read-Only)
- Read the blog post — Detect format (MDX, markdown, HTML)
- Run the quality checklist against
../blog/references/quality-scoring.md:
- Count fabricated vs sourced statistics
- Check answer-first formatting (H2 -> stat in first sentence?)
- Count images and charts (type diversity?)
- Measure paragraph lengths (any > 150 words?)
- Check heading hierarchy (H1 -> H2 -> H3, no skips?)
- Look for FAQ schema
- Check freshness signals (lastUpdated, dateModified)
- Assess self-promotion level
- Evaluate citation tier quality
- AI content detection scan:
- Burstiness score — Measure sentence length variance across the post. Low
variance (most sentences within 3-5 words of each other) is a strong AI signal.
Calculate: standard deviation of sentence word counts. Target SD > 6.
- Known AI phrase scan — Check for these high-frequency AI phrases:
- "in today's digital landscape", "it's important to note", "dive into"
- "game-changer", "navigate the landscape", "revolutionize", "seamlessly"
- "cutting-edge", "harness the power of", "leverage" (as verb)
- "delve", "crucial", "elevate", "foster", "landscape" (overused)
- "multifaceted", "robust", "tapestry", "embark"
- Full list in
agents/blog-writer.md
- Vocabulary diversity — Calculate Type-Token Ratio (TTR): unique words /
total words. Low TTR (< 0.40) suggests AI-generated repetitive phrasing.
Target TTR > 0.50 for natural prose.
- AI content percentage estimate — Based on burstiness, phrase density, and
TTR, estimate what percentage of the content reads as AI-generated (0-100%).
Report as: "AI content estimate: ~X%"
- Cannibalization check:
- Identify the post's primary keyword from title, H1, and first paragraph
- Search the blog directory for other posts targeting the same keyword:
- Grep headings and meta descriptions across all blog posts
- Flag any posts with significant keyword overlap
- If cannibalization found, report:
- Which posts compete for the same keyword
- Recommend: merge (combine into one stronger post) or differentiate
(shift one post to a related but distinct keyword)
- Calculate current score across 5 categories:
- Score across 5 categories (Content Quality 30, SEO Optimization 25, E-E-A-T Signals 15, Technical Elements 15, AI Citation Readiness 15)
- Total: 0-100
- Present audit summary with specific findings, AI detection results, cannibalization status, and score
- Enter plan mode — Present section-by-section optimization plan
Wait for user approval before proceeding.
Phase 2: Research
- Identify the blog's core topic from existing content
- Find replacement statistics for any fabricated/unsourced data:
- Search:
[topic] study 2025 2026 data statistics
- Target tier 1-3 sources only
- Find images if post has fewer than 3:
- Pixabay:
site:pixabay.com [topic keywords]
- Unsplash:
site:unsplash.com [topic keywords]
- Verify each URL returns HTTP 200
- Plan charts if post has fewer than 2:
- Identify data suitable for visualization
- Select diverse chart types
Phase 3: Chart Generation (Built-In)
When the post needs more visual elements, invoke the blog-chart sub-skill:
- Select chart type using the diversity rule (no repeated types per post)
- Pass: chart type, title, data values, source, platform format
- Embed the returned SVG directly within a
<figure> wrapper
- Target 2-4 charts per 2,000-word post
See ../blog/references/visual-media.md for chart type selection and styling rules.
Phase 4: Content Rewrite
Apply changes in this order:
4a. Preserve What Works
- Keep the author's voice and unique perspective
- Preserve original insights and first-hand experience
- Keep existing quality images and charts
- Maintain internal links
4b. Fix Frontmatter
- Add
lastUpdated: "YYYY-MM-DD" (today's date)
- Keep original
date unchanged
- Fix meta description: fact-dense, 150-160 chars, includes 1 statistic
- Add
coverImage + coverImageAlt + ogImage if missing
- Search Pixabay/Unsplash/Pexels for wide hero image (1200x630)
- Or generate custom SVG cover via
blog-chart (text-on-gradient with key stat)
- Verify tags/categories are appropriate
4c. Apply Answer-First Formatting
Every H2 section MUST open with a 40-60 word paragraph containing:
- At least one specific statistic with source attribution
- A direct answer to the heading's implicit question
4d. Replace Fabricated Statistics
- Search for patterns: "X% of...", "X out of Y...", unsourced claims
- Replace with real data from tier 1-3 sources
- Always include inline attribution:
([Source Name](url), year)
4e. Improve Headings
- Convert statement headings to questions where natural (60-70% target)
- Keep 2-3 statement headings for variety
- Ensure keyword appears in 2-3 headings naturally
4f. Fix Paragraph Length
- Split any paragraph > 150 words
- Target 40-80 words per paragraph
- Ensure each paragraph starts with its most important sentence
4g. Add Visual Elements
- Embed new images after H2 headings, spaced evenly
- Embed charts within relevant sections
- Adapt embed format to detected platform (MDX vs markdown vs HTML)
4h. Add/Improve FAQ
- If no FAQ exists, add one (3-5 questions)
- If FAQ exists, ensure answers are 40-60 words with statistics
- Add FAQ schema markup appropriate to platform
4i. Reduce Self-Promotion
- Max 1 brand mention (author bio context only)
- Remove "At [Company], we..." patterns
- Convert promotional sections to educational content
4j. Citation Capsule Injection
For each H2 section, generate (or improve existing) a citation capsule:
- 40-60 word self-contained passage per H2
- Contains: one specific claim + one data point + source attribution
- Written in a declarative style so an AI system could extract and quote it directly
- Placed naturally within the section body, not as a separate callout
Example:
According to a 2026 Gartner study, 58% of enterprise buyers now consult AI
assistants before contacting a vendor ([Gartner](https://www.gartner.com), 2026).
This shift means B2B content must answer specific questions concisely enough
for AI systems to extract and cite in their responses.
Capsules map to the "AI Citation Readiness" category (15 points) in
../blog/references/quality-scoring.md.
4k. Anti-AI-Detection Patterns
Apply these transformations to reduce AI-detectable writing patterns:
- Replace flagged phrases — Swap every detected AI phrase (from the scan in
Phase 1 step 3) with a natural alternative. Examples:
- "it's important to note" -> "worth noting" or "keep in mind"
- "in today's digital landscape" -> "right now" or "in [specific year]"
- "leverage" -> "use", "apply", "take advantage of"
- "delve" -> "look at", "explore", "dig into"
- "robust" -> "strong", "solid", "reliable"
- "crucial" -> "key", "essential", "critical" (or restructure the sentence)
- Vary sentence length deliberately — After rewriting, scan each paragraph.
Inject short punchy sentences (5-10 words) between longer ones (18-25 words).
Target: no more than 3 consecutive sentences within 5 words of each other's length.
- Inject rhetorical questions — Add at least one rhetorical question every
200-300 words to break up declarative monotony.
- Use contractions naturally — Replace formal constructions with contractions
where they sound natural: "it is" -> "it's", "we have" -> "we've",
"do not" -> "don't", "is not" -> "isn't".
- Include hedging language — Sprinkle first-person hedges that signal real
experience: "in our experience", "we've found that", "from what we've seen",
"this tends to", "it depends on".
4l. TL;DR Box
If the post lacks a TL;DR box, add one immediately after the introduction:
> **TL;DR:** [40-60 word standalone summary. Contains the key finding or
> recommendation plus 1 statistic with source. Self-contained — reader gets
> the core value without reading the full article.]
If a TL;DR already exists, verify it meets the 40-60 word requirement and
contains a statistic with source attribution.
4m. Information Gain Marker Injection
Review the post for original value and tag it:
[ORIGINAL DATA] — Any proprietary data, survey results, experiments, or
case study metrics the author collected first-hand
[PERSONAL EXPERIENCE] — First-hand observations, lessons learned
[UNIQUE INSIGHT] — Novel analysis, contrarian perspectives backed by data
If the post lacks original value markers:
- Ask the author for first-hand data or experience to include
- At minimum, add analytical insights that connect existing research in new ways
- Target: at least 2-3 markers per post
Use HTML comments (<!-- [ORIGINAL DATA] -->) or visible callouts depending
on the post's style.
Phase 5: Verification
After rewriting, verify all quality gates pass:
Core Quality Gates
- Every H2 opens with a statistic + source
- No paragraph exceeds 150 words
- Zero fabricated statistics
- Heading hierarchy is clean
- FAQ section present with schema
- Images have descriptive alt text
- Cover image present in frontmatter (coverImage + ogImage)
- If MDX: build the project to verify no compilation errors
New Element Verification
- TL;DR box present after introduction (40-60 words, contains statistic)
- At least 2-3 information gain markers present
- Citation capsules in major H2 sections (40-60 words, self-contained)
- Internal linking zones marked or actual links present (5-10 per 2,000 words)
- No AI-detectable phrases remain from banned list
Burstiness and Naturalness Check
- Sentence length variance: SD > 6 (mix of short and long sentences)
- Contractions used naturally throughout
- Rhetorical questions present (1 per 200-300 words)
- AI content estimate reduced from audit baseline
- Score improved across all 5 categories vs Phase 1 audit
Phase 6: Summary
## Blog Optimization Complete: [Title]
### Score Change
- Before: [X]/100 ([Rating])
- Content Quality: [X]/30
- SEO Optimization: [X]/25
- E-E-A-T Signals: [X]/15
- Technical Elements: [X]/15
- AI Citation Readiness: [X]/15
- After: [Y]/100 ([Rating])
- Content Quality: [Y]/30
- SEO Optimization: [Y]/25
- E-E-A-T Signals: [Y]/15
- Technical Elements: [Y]/15
- AI Citation Readiness: [Y]/15
### AI Detection
- Before: ~[X]% AI-detected content
- After: ~[Y]% AI-detected content
- Phrases replaced: [N]
- Burstiness improved: [before SD] -> [after SD]
### Cannibalization
- [Status: none found / flagged N posts / resolved]
### Changes Made
- [X] statistics replaced with sourced data
- [X] SVG charts added (types: ...)
- [X] images added from Pixabay/Unsplash
- Answer-first formatting applied to [N] H2 sections
- FAQ schema injected with [N] questions
- TL;DR box: [added/updated]
- Information gain markers: [N] ([types])
- Citation capsules: [N] across H2 sections
- AI phrases replaced: [N]
- lastUpdated set to [date]
- Self-promotion reduced to [N] mentions
### Visual Elements
- Charts: [count] ([types])
- Images: [count]
### Ready for
- `$blog analyze <file>` to verify final score
- Publishing / deploying
Update Mode
When invoked as $blog update <file>, focus on freshness:
- Update statistics to latest available data (2025-2026)
- Add new developments since last update
- Refresh images if older than 1 year
- Update
lastUpdated in frontmatter
- Preserve the existing structure — minimize rewrites
- Target: at least 30% content change to register as "fresh" for AI crawlers
1---2name: blog-rewrite3description: >4 Rewrite and optimize existing blog posts for Google rankings (December 20255 Core Update, E-E-A-T) and AI citations (GEO/AEO). Replaces fabricated statistics6 with sourced data, applies answer-first formatting, adds Pixabay/Unsplash7 images, generates built-in SVG charts, injects FAQ schema, performs8 AI content detection, adds citation capsules and information gain markers,9 and updates freshness signals. Works with any blog format (MDX, markdown, HTML).10 Use when user says "rewrite blog", "optimize blog", "update blog",11 "improve blog", "fix blog", "refresh blog post", "blog optimization".12metadata:13 source: https://skillmd.com/skills/megastep/blog-rewrite14---1516# Blog Rewriter -- Optimize Existing Posts1718Rewrites and optimizes existing blog posts for dual ranking: Google search19and AI citation platforms. Preserves the author's voice while applying the206 pillars of optimization.2122**Key references:**23- `../blog/references/quality-scoring.md` — 5-category scoring (Content 30, SEO 25, E-E-A-T 15, Technical 15, AI Citation 15)24- `../blog/references/eeat-signals.md` — Experience, expertise, authority, trust markers25- `../blog/references/internal-linking.md` — Linking strategy and anchor text rules26- `../blog/references/visual-media.md` — Image sourcing and chart styling2728## Workflow2930### Phase 1: Audit (Read-Only)31321. **Read the blog post** — Detect format (MDX, markdown, HTML)332. **Run the quality checklist** against `../blog/references/quality-scoring.md`:34 - Count fabricated vs sourced statistics35 - Check answer-first formatting (H2 -> stat in first sentence?)36 - Count images and charts (type diversity?)37 - Measure paragraph lengths (any > 150 words?)38 - Check heading hierarchy (H1 -> H2 -> H3, no skips?)39 - Look for FAQ schema40 - Check freshness signals (lastUpdated, dateModified)41 - Assess self-promotion level42 - Evaluate citation tier quality433. **AI content detection scan**:44 - **Burstiness score** — Measure sentence length variance across the post. Low45 variance (most sentences within 3-5 words of each other) is a strong AI signal.46 Calculate: standard deviation of sentence word counts. Target SD > 6.47 - **Known AI phrase scan** — Check for these high-frequency AI phrases:48 - "in today's digital landscape", "it's important to note", "dive into"49 - "game-changer", "navigate the landscape", "revolutionize", "seamlessly"50 - "cutting-edge", "harness the power of", "leverage" (as verb)51 - "delve", "crucial", "elevate", "foster", "landscape" (overused)52 - "multifaceted", "robust", "tapestry", "embark"53 - Full list in `agents/blog-writer.md`54 - **Vocabulary diversity** — Calculate Type-Token Ratio (TTR): unique words /55 total words. Low TTR (< 0.40) suggests AI-generated repetitive phrasing.56 Target TTR > 0.50 for natural prose.57 - **AI content percentage estimate** — Based on burstiness, phrase density, and58 TTR, estimate what percentage of the content reads as AI-generated (0-100%).59 Report as: "AI content estimate: ~X%"604. **Cannibalization check**:61 - Identify the post's primary keyword from title, H1, and first paragraph62 - Search the blog directory for other posts targeting the same keyword:63 - Grep headings and meta descriptions across all blog posts64 - Flag any posts with significant keyword overlap65 - If cannibalization found, report:66 - Which posts compete for the same keyword67 - Recommend: **merge** (combine into one stronger post) or **differentiate**68 (shift one post to a related but distinct keyword)695. **Calculate current score** across 5 categories:70 - Score across 5 categories (Content Quality 30, SEO Optimization 25, E-E-A-T Signals 15, Technical Elements 15, AI Citation Readiness 15)71 - Total: 0-100726. **Present audit summary** with specific findings, AI detection results, cannibalization status, and score737. **Enter plan mode** — Present section-by-section optimization plan7475Wait for user approval before proceeding.7677### Phase 2: Research78791. **Identify the blog's core topic** from existing content802. **Find replacement statistics** for any fabricated/unsourced data:81 - Search: `[topic] study 2025 2026 data statistics`82 - Target tier 1-3 sources only833. **Find images** if post has fewer than 3:84 - Pixabay: `site:pixabay.com [topic keywords]`85 - Unsplash: `site:unsplash.com [topic keywords]`86 - Verify each URL returns HTTP 200874. **Plan charts** if post has fewer than 2:88 - Identify data suitable for visualization89 - Select diverse chart types9091### Phase 3: Chart Generation (Built-In)9293When the post needs more visual elements, invoke the `blog-chart` sub-skill:94951. Select chart type using the diversity rule (no repeated types per post)962. Pass: chart type, title, data values, source, platform format973. Embed the returned SVG directly within a `<figure>` wrapper984. Target 2-4 charts per 2,000-word post99100See `../blog/references/visual-media.md` for chart type selection and styling rules.101102### Phase 4: Content Rewrite103104Apply changes in this order:105106#### 4a. Preserve What Works107- Keep the author's voice and unique perspective108- Preserve original insights and first-hand experience109- Keep existing quality images and charts110- Maintain internal links111112#### 4b. Fix Frontmatter113- Add `lastUpdated: "YYYY-MM-DD"` (today's date)114- Keep original `date` unchanged115- Fix meta description: fact-dense, 150-160 chars, includes 1 statistic116- Add `coverImage` + `coverImageAlt` + `ogImage` if missing117 - Search Pixabay/Unsplash/Pexels for wide hero image (1200x630)118 - Or generate custom SVG cover via `blog-chart` (text-on-gradient with key stat)119- Verify tags/categories are appropriate120121#### 4c. Apply Answer-First Formatting122Every H2 section MUST open with a 40-60 word paragraph containing:123- At least one specific statistic with source attribution124- A direct answer to the heading's implicit question125126#### 4d. Replace Fabricated Statistics127- Search for patterns: "X% of...", "X out of Y...", unsourced claims128- Replace with real data from tier 1-3 sources129- Always include inline attribution: `([Source Name](url), year)`130131#### 4e. Improve Headings132- Convert statement headings to questions where natural (60-70% target)133- Keep 2-3 statement headings for variety134- Ensure keyword appears in 2-3 headings naturally135136#### 4f. Fix Paragraph Length137- Split any paragraph > 150 words138- Target 40-80 words per paragraph139- Ensure each paragraph starts with its most important sentence140141#### 4g. Add Visual Elements142- Embed new images after H2 headings, spaced evenly143- Embed charts within relevant sections144- Adapt embed format to detected platform (MDX vs markdown vs HTML)145146#### 4h. Add/Improve FAQ147- If no FAQ exists, add one (3-5 questions)148- If FAQ exists, ensure answers are 40-60 words with statistics149- Add FAQ schema markup appropriate to platform150151#### 4i. Reduce Self-Promotion152- Max 1 brand mention (author bio context only)153- Remove "At [Company], we..." patterns154- Convert promotional sections to educational content155156#### 4j. Citation Capsule Injection157For each H2 section, generate (or improve existing) a citation capsule:158- 40-60 word self-contained passage per H2159- Contains: one specific claim + one data point + source attribution160- Written in a declarative style so an AI system could extract and quote it directly161- Placed naturally within the section body, not as a separate callout162163Example:164```markdown165According to a 2026 Gartner study, 58% of enterprise buyers now consult AI166assistants before contacting a vendor ([Gartner](https://www.gartner.com), 2026).167This shift means B2B content must answer specific questions concisely enough168for AI systems to extract and cite in their responses.169```170171Capsules map to the "AI Citation Readiness" category (15 points) in172`../blog/references/quality-scoring.md`.173174#### 4k. Anti-AI-Detection Patterns175Apply these transformations to reduce AI-detectable writing patterns:176- **Replace flagged phrases** — Swap every detected AI phrase (from the scan in177 Phase 1 step 3) with a natural alternative. Examples:178 - "it's important to note" -> "worth noting" or "keep in mind"179 - "in today's digital landscape" -> "right now" or "in [specific year]"180 - "leverage" -> "use", "apply", "take advantage of"181 - "delve" -> "look at", "explore", "dig into"182 - "robust" -> "strong", "solid", "reliable"183 - "crucial" -> "key", "essential", "critical" (or restructure the sentence)184- **Vary sentence length deliberately** — After rewriting, scan each paragraph.185 Inject short punchy sentences (5-10 words) between longer ones (18-25 words).186 Target: no more than 3 consecutive sentences within 5 words of each other's length.187- **Inject rhetorical questions** — Add at least one rhetorical question every188 200-300 words to break up declarative monotony.189- **Use contractions naturally** — Replace formal constructions with contractions190 where they sound natural: "it is" -> "it's", "we have" -> "we've",191 "do not" -> "don't", "is not" -> "isn't".192- **Include hedging language** — Sprinkle first-person hedges that signal real193 experience: "in our experience", "we've found that", "from what we've seen",194 "this tends to", "it depends on".195196#### 4l. TL;DR Box197If the post lacks a TL;DR box, add one immediately after the introduction:198```markdown199> **TL;DR:** [40-60 word standalone summary. Contains the key finding or200> recommendation plus 1 statistic with source. Self-contained — reader gets201> the core value without reading the full article.]202```203If a TL;DR already exists, verify it meets the 40-60 word requirement and204contains a statistic with source attribution.205206#### 4m. Information Gain Marker Injection207Review the post for original value and tag it:208- `[ORIGINAL DATA]` — Any proprietary data, survey results, experiments, or209 case study metrics the author collected first-hand210- `[PERSONAL EXPERIENCE]` — First-hand observations, lessons learned211- `[UNIQUE INSIGHT]` — Novel analysis, contrarian perspectives backed by data212213If the post lacks original value markers:214- Ask the author for first-hand data or experience to include215- At minimum, add analytical insights that connect existing research in new ways216- Target: at least 2-3 markers per post217218Use HTML comments (`<!-- [ORIGINAL DATA] -->`) or visible callouts depending219on the post's style.220221### Phase 5: Verification222223After rewriting, verify all quality gates pass:224225#### Core Quality Gates2261. Every H2 opens with a statistic + source2272. No paragraph exceeds 150 words2283. Zero fabricated statistics2294. Heading hierarchy is clean2305. FAQ section present with schema2316. Images have descriptive alt text2327. Cover image present in frontmatter (coverImage + ogImage)2338. If MDX: build the project to verify no compilation errors234235#### New Element Verification2369. TL;DR box present after introduction (40-60 words, contains statistic)23710. At least 2-3 information gain markers present23811. Citation capsules in major H2 sections (40-60 words, self-contained)23912. Internal linking zones marked or actual links present (5-10 per 2,000 words)24013. No AI-detectable phrases remain from banned list241242#### Burstiness and Naturalness Check24314. Sentence length variance: SD > 6 (mix of short and long sentences)24415. Contractions used naturally throughout24516. Rhetorical questions present (1 per 200-300 words)24617. AI content estimate reduced from audit baseline24718. Score improved across all 5 categories vs Phase 1 audit248249### Phase 6: Summary250251```252## Blog Optimization Complete: [Title]253254### Score Change255- Before: [X]/100 ([Rating])256 - Content Quality: [X]/30257 - SEO Optimization: [X]/25258 - E-E-A-T Signals: [X]/15259 - Technical Elements: [X]/15260 - AI Citation Readiness: [X]/15261- After: [Y]/100 ([Rating])262 - Content Quality: [Y]/30263 - SEO Optimization: [Y]/25264 - E-E-A-T Signals: [Y]/15265 - Technical Elements: [Y]/15266 - AI Citation Readiness: [Y]/15267268### AI Detection269- Before: ~[X]% AI-detected content270- After: ~[Y]% AI-detected content271- Phrases replaced: [N]272- Burstiness improved: [before SD] -> [after SD]273274### Cannibalization275- [Status: none found / flagged N posts / resolved]276277### Changes Made278- [X] statistics replaced with sourced data279- [X] SVG charts added (types: ...)280- [X] images added from Pixabay/Unsplash281- Answer-first formatting applied to [N] H2 sections282- FAQ schema injected with [N] questions283- TL;DR box: [added/updated]284- Information gain markers: [N] ([types])285- Citation capsules: [N] across H2 sections286- AI phrases replaced: [N]287- lastUpdated set to [date]288- Self-promotion reduced to [N] mentions289290### Visual Elements291- Charts: [count] ([types])292- Images: [count]293294### Ready for295- `$blog analyze <file>` to verify final score296- Publishing / deploying297```298299## Update Mode300301When invoked as `$blog update <file>`, focus on freshness:3021. Update statistics to latest available data (2025-2026)3032. Add new developments since last update3043. Refresh images if older than 1 year3054. Update `lastUpdated` in frontmatter3065. Preserve the existing structure — minimize rewrites3076. Target: at least 30% content change to register as "fresh" for AI crawlers