[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6C2KA2MKWA4RR4750F554P3":3,"skill-same:01M6C2KA2MKWA4RR4750F554P3":115},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":34,"publishedAt":61,"linkOnly":62,"markdownContent":63,"markdownBody":64,"frontMatter":65,"refName":73,"commitSha":74,"skillDirectory":75,"forksCount":14,"repositoryHtmlUrl":76,"repositoryHomepageUrl":6,"ownerHtmlUrl":77,"ownerBlogUrl":6,"ownerDescription":78,"updatedAt":79,"install":80,"resources":92},"01M6C2KA2MKWA4RR4750F554P3","seo-cluster",null,"SERP-based semantic topic clustering for content architecture planning. Groups keywords by actual Google SERP overlap (not text similarity), designs hub-and-spoke content clusters with internal link matrices, and generates interactive visualizations. Optionally executes content creation if claude-blog is installed. Use when user says \"topic cluster\", \"content cluster\", \"semantic clustering\", \"pillar page\", \"hub and spoke\", \"content architecture\", \"keyword grouping\", or \"cluster plan\".","noirelab\u002Flinux-setup","noirelab","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F142925921?v=4","01M6C29R3FEDGFAXAT0KXY28QX","Kaíque M. Lima","01M6C29R811WX04B0GYHTH1XV6",0,"MIT","https:\u002F\u002Fgithub.com\u002Fnoirelab\u002Flinux-setup\u002Fblob\u002Fmain\u002Fconfigs\u002Fcommon\u002Fclaude\u002Fskills\u002Fseo-cluster\u002FSKILL.md",[18,23,26,31],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000018","写作与文案","SCENARIO","使用场景",{"categoryNo":24,"categoryName":25,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000019","营销推广",{"categoryNo":27,"categoryName":28,"taxonomyType":29,"taxonomyTypeDesc":30},"01JBSEEDCAT000000000000035","市场营销","PROFESSION","适用职业",{"categoryNo":32,"categoryName":33,"taxonomyType":29,"taxonomyTypeDesc":30},"01JBSEEDCAT000000000000036","内容创作者",[35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60],"seo","cluster","seo cluster","interactive","Clusters","when","hub-and-spoke","URL","creation","seed","content creation","grouping","use","user","google","challenge","page","text","blog","plan","miller","visualizations","clustering","content","planning","architecture","2026-11-10 10:12:30",false,"---\nname: seo-cluster\ndescription: >\n  SERP-based semantic topic clustering for content architecture planning. Groups\n  keywords by actual Google SERP overlap (not text similarity), designs hub-and-spoke\n  content clusters with internal link matrices, and generates interactive\n  visualizations. Optionally executes content creation if claude-blog is installed.\n  Use when user says \"topic cluster\", \"content cluster\", \"semantic clustering\",\n  \"pillar page\", \"hub and spoke\", \"content architecture\", \"keyword grouping\",\n  or \"cluster plan\".\nuser-invocable: true\nargument-hint: \"\u003Cseed-keyword or url>\"\nlicense: MIT\nmetadata:\n  author: AgriciDaniel\n  original_author: \"Lutfiya Miller (Pro Hub Challenge Winner)\"\n  version: \"2.2.4\"\n  category: seo\n---\n\n# Semantic Topic Clustering\n\nSERP-overlap-driven keyword clustering for content architecture. Groups keywords\nby how Google actually ranks them (shared top-10 results), not by text similarity.\nDesigns hub-and-spoke content clusters with internal link matrices and generates\ninteractive cluster map visualizations.\n\n**Scripts:** Located at the plugin root `scripts\u002F` directory.\n\n---\n\n## Quick Reference\n\n| Command | What it does |\n|---------|-------------|\n| `\u002Fseo cluster plan \u003Cseed-keyword>` | Full planning workflow: expand, cluster, architect, visualize |\n| `\u002Fseo cluster plan --from strategy` | Import from existing `\u002Fseo plan` output |\n| `\u002Fseo cluster execute` | Execute plan: create content via claude-blog or output briefs |\n| `\u002Fseo cluster map` | Regenerate the interactive cluster visualization |\n\n---\n\n## Planning Workflow\n\n### Step 1: Seed Keyword Expansion\n\nExpand the seed keyword into 30-50 variants using WebSearch:\n\n1. **Related searches**: Search the seed, extract \"related searches\" and \"people also search for\"\n2. **People Also Ask (PAA)**: Extract all PAA questions from SERP results\n3. **Long-tail modifiers**: Append common modifiers: \"best\", \"how to\", \"vs\", \"for beginners\", \"tools\", \"examples\", \"guide\", \"template\", \"mistakes\", \"checklist\"\n4. **Question mining**: Generate who\u002Fwhat\u002Fwhen\u002Fwhere\u002Fwhy\u002Fhow variants\n5. **Intent modifiers**: Add commercial modifiers: \"pricing\", \"review\", \"alternative\", \"comparison\", \"free\", \"top\"\n\n**Deduplication:** Normalize variants (lowercase, strip articles), remove exact duplicates.\nTarget: 30-50 unique keyword variants. If under 30, run a second expansion pass\nwith the top PAA questions as seeds.\n\n### Step 2: SERP Overlap Clustering\n\nThis is the core differentiator. Load `references\u002Fserp-overlap-methodology.md` for\nthe full algorithm.\n\n**Process:**\n1. Group keywords by initial intent guess (reduces pairwise comparisons)\n2. For each candidate pair within a group, WebSearch both keywords\n3. Count shared URLs in the top 10 organic results (ignore ads, featured snippets, PAA)\n4. Apply thresholds:\n\n| Shared Results | Relationship | Action |\n|---------------|-------------|--------|\n| 7-10 | Same post | Merge into single target page |\n| 4-6 | Same cluster | Group under same spoke cluster |\n| 2-3 | Interlink | Place in adjacent clusters, add cross-links |\n| 0-1 | Separate | Assign to different clusters or exclude |\n\n**Optimization:** With 40 keywords, full pairwise = 780 comparisons. Instead:\n- Pre-group by intent (4 groups of ~10 = 4 x 45 = 180 comparisons)\n- Only cross-check group boundary keywords\n- Skip pairs where both are long-tail variants of the same head term (assume same cluster)\n\n**DataForSEO integration:** If DataForSEO MCP is available, use `serp_organic_live_advanced`\ninstead of WebSearch for SERP data. Run `\u002Fhome\u002Fnoirelab\u002F.config\u002Fopencode\u002Fskills\u002Fseo\u002Fbin\u002Fclaude-seo run dataforseo_costs.py check serp_organic_live_advanced --count N`\nbefore each batch. If `\"status\": \"needs_approval\"`, show cost estimate and ask user.\nIf `\"status\": \"blocked\"`, fall back to WebSearch.\n\n### Step 3: Intent Classification\n\nClassify each keyword into one of four intent categories:\n\n| Intent | Signals | Include in Clusters? |\n|--------|---------|---------------------|\n| Informational | how, what, why, guide, tutorial, learn | Yes |\n| Commercial | best, top, review, comparison, vs, alternative | Yes |\n| Transactional | buy, price, discount, coupon, order, sign up | Yes |\n| Navigational | brand names, specific product names, login | No (exclude) |\n\nRemove navigational keywords from clustering. Flag borderline cases for\nmanual review. Keywords can have mixed intent (e.g., \"best CRM software\" is\nboth commercial and informational) -- classify by dominant intent.\n\n### Step 4: Hub-and-Spoke Architecture\n\nLoad `references\u002Fhub-spoke-architecture.md` for full specifications.\n\n**Design the cluster structure:**\n\n1. **Select the pillar keyword**: Highest volume, broadest intent, most SERP overlap with other keywords\n2. **Group spokes into clusters**: Each cluster is a subtopic area (2-5 clusters per pillar)\n3. **Assign posts to clusters**: Each cluster gets 2-4 spoke posts\n4. **Select templates per post**: Based on intent classification:\n\n| Intent Pattern | Template Options |\n|---------------|-----------------|\n| Informational (broad) | ultimate-guide |\n| Informational (how) | how-to |\n| Informational (list) | listicle |\n| Informational (concept) | explainer |\n| Commercial (compare) | comparison |\n| Commercial (evaluate) | review |\n| Commercial (rank) | best-of |\n| Transactional | landing-page |\n\n5. **Set word count targets:**\n   - Pillar page: 2500-4000 words\n   - Spoke posts: 1200-1800 words\n\n6. **Cannibalization check**: No two posts share the same primary keyword. If SERP\n   overlap is 7+, merge those keywords into a single post targeting both.\n\n### Step 5: Internal Link Matrix\n\nDesign the bidirectional linking structure:\n\n| Link Type | Direction | Requirement |\n|-----------|-----------|-------------|\n| Spoke to pillar | spoke -> pillar | Mandatory (every spoke) |\n| Pillar to spoke | pillar -> spoke | Mandatory (every spoke) |\n| Spoke to spoke (within cluster) | spoke \u003C-> spoke | 2-3 links per post |\n| Cross-cluster | spoke -> spoke (other cluster) | 0-1 links per post |\n\n**Rules:**\n- Every post must have minimum 3 incoming internal links\n- No orphan pages (every post reachable from pillar in 2 clicks)\n- Anchor text must use target keyword or close variant (no \"click here\")\n- Link placement: within body content, not just navigation\u002Fsidebar\n\nGenerate the link matrix as a JSON adjacency list:\n```json\n{\n  \"links\": [\n    { \"from\": \"pillar\", \"to\": \"cluster-0-post-0\", \"type\": \"mandatory\", \"anchor\": \"keyword\" },\n    { \"from\": \"cluster-0-post-0\", \"to\": \"pillar\", \"type\": \"mandatory\", \"anchor\": \"keyword\" }\n  ]\n}\n```\n\n### Step 6: Interactive Cluster Map\n\nGenerate `cluster-map.html` using the template at `templates\u002Fcluster-map.html`.\n\n1. Read the template file\n2. Build the `CLUSTER_DATA` JSON object from the cluster plan:\n   ```javascript\n   {\n     pillar: { title, keyword, volume, template, wordCount, url },\n     clusters: [{ name, color, posts: [{ title, keyword, volume, template, wordCount, url, status }] }],\n     links: [{ from, to, type }],\n     meta: { totalPosts, totalClusters, totalLinks, estimatedWords }\n   }\n   ```\n3. Replace the `CLUSTER_DATA` placeholder in the template with the actual JSON\n4. Write the completed HTML file to the output directory\n5. Inform user: \"Open `cluster-map.html` in a browser to explore the interactive cluster map.\"\n\n---\n\n## Strategy Import\n\nWhen invoked with `--from strategy`:\n\n1. Look for the most recent `\u002Fseo plan` output in the current directory (search for\n   files matching `*SEO*Plan*`, `*strategy*`, `*content-strategy*`)\n2. Parse markdown tables for: keywords, page types, content pillars, URL structures\n3. Validate extracted data: check for duplicates, missing keywords, incomplete entries\n4. Enrich with SERP data: run SERP overlap analysis on extracted keywords\n5. Build cluster plan using the imported keywords as the starting set (skip Step 1)\n\nIf no strategy file is found, prompt the user: \"No existing SEO plan found in the\ncurrent directory. Run `\u002Fseo plan` first, or provide a seed keyword for fresh clustering.\"\n\n---\n\n## Execution Workflow\n\nWhen `\u002Fseo cluster execute` is invoked:\n\n### Check for claude-blog\n\n```\nTest: Does ~\u002F.claude\u002Fskills\u002Fblog\u002FSKILL.md exist?\n```\n\n**If claude-blog IS installed:**\n\n1. Load `references\u002Fexecution-workflow.md` for the full algorithm\n2. Read `cluster-plan.json` from the current directory\n3. Check for resume state: scan output directory for already-written posts\n4. Execute in priority order: pillar first, then spokes by volume (highest first)\n5. For each post, invoke the `blog-write` skill with cluster context:\n   - Cluster role (pillar or spoke)\n   - Position in cluster (cluster index, post index)\n   - Target keyword and secondary keywords\n   - Template type and word count target\n   - Internal links to include (with anchors)\n   - Links to receive from future posts (placeholder markers)\n6. After each post is written, scan previous posts for backward link placeholders\n   and inject the new post's URL\n7. After all posts are written, generate the cluster scorecard\n\n**If claude-blog is NOT installed:**\n\n1. Generate detailed content briefs for each post in the cluster plan\n2. Each brief includes:\n   - Title and meta description\n   - Primary keyword and secondary keywords\n   - Template type and suggested structure (H2\u002FH3 outline)\n   - Word count target\n   - Internal links to include (with anchor text)\n   - Key points to cover\n   - Competing pages to differentiate from\n3. Write briefs to `cluster-briefs\u002F` directory as individual markdown files\n4. Inform user: \"Install [claude-blog](https:\u002F\u002Fgithub.com\u002FAgriciDaniel\u002Fclaude-blog)\n   to auto-create content. Briefs saved to `cluster-briefs\u002F`.\"\n\n---\n\n## Cluster Scorecard\n\nPost-execution quality report. Run automatically after `\u002Fseo cluster execute` or\non demand via analysis of the output directory.\n\n| Metric | Target | How Measured |\n|--------|--------|-------------|\n| Coverage | 100% | Posts written \u002F posts planned |\n| Link Density | 3+ per post | Count internal links per post |\n| Orphan Pages | 0 | Posts with \u003C 1 incoming link |\n| Cannibalization | 0 conflicts | Check for duplicate primary keywords |\n| Image Count | 1+ per post | Posts with at least one image |\n| Pillar Links | 100% | All spokes link to pillar and vice versa |\n| Cross-Links | 80%+ | Recommended spoke-to-spoke links implemented |\n| Content Gaps | 0 | Planned posts that were skipped or incomplete |\n\n---\n\n## Map Regeneration\n\nWhen `\u002Fseo cluster map` is invoked:\n\n1. Read `cluster-plan.json` from the current directory\n2. Scan output directory and update post statuses (planned vs written)\n3. Regenerate `cluster-map.html` with updated statuses\n4. Report: posts written vs planned, link completion percentage\n\n---\n\n## Output Files\n\nAll outputs are written to the current working directory:\n\n| File | Description |\n|------|-------------|\n| `cluster-plan.json` | Machine-readable cluster plan (full data) |\n| `cluster-plan.md` | Human-readable cluster plan summary |\n| `cluster-map.html` | Interactive SVG visualization |\n| `cluster-briefs\u002F` | Content briefs (if no claude-blog) |\n| `cluster-scorecard.md` | Post-execution quality report |\n\n---\n\n## Cross-Skill Integration\n\n| Skill | Relationship |\n|-------|-------------|\n| `seo-plan` | Import source: strategy import reads seo-plan output |\n| `seo-content` | Quality check: E-E-A-T validation of generated content |\n| `seo-schema` | Schema markup: Article, BreadcrumbList, ItemList for cluster pages |\n| `seo-dataforseo` | Data source: SERP data when DataForSEO MCP is available |\n| `seo-google` | Reporting: generate PDF report of cluster plan and scorecard |\n\nAfter cluster planning or execution completes, offer:\n\"Generate a PDF report? Use `\u002Fseo google report`\"\n\n---\n\n## Error Handling\n\n| Error | Cause | Resolution |\n|-------|-------|------------|\n| \"No seed keyword provided\" | Missing argument | Prompt user for seed keyword or URL |\n| \"Insufficient keyword variants\" | Expansion yielded \u003C 15 keywords | Run second expansion pass with PAA questions |\n| \"SERP data unavailable\" | WebSearch and DataForSEO both failing | Retry after 30s; if persistent, use intent-only clustering with warning |\n| \"No strategy file found\" | `--from strategy` but no plan exists | Prompt user to run `\u002Fseo plan` first |\n| \"cluster-plan.json not found\" | Execute without planning | Prompt user to run `\u002Fseo cluster plan` first |\n| \"claude-blog not installed\" | Execute attempted without blog skill | Generate content briefs instead; suggest installation |\n| \"DataForSEO budget exceeded\" | Cost check returned \"blocked\" | Fall back to WebSearch; inform user |\n| \"Duplicate primary keywords\" | Cannibalization detected | Merge affected posts or reassign keywords |\n| \"Orphan page detected\" | Post missing incoming links | Add links from nearest cluster siblings |\n| \"Resume state corrupted\" | Mismatch between plan and output | Rebuild state from output directory scan |\n\n---\n\n## Security\n\n- All URLs fetched via `\u002Fhome\u002Fnoirelab\u002F.config\u002Fopencode\u002Fskills\u002Fseo\u002Fbin\u002Fclaude-seo run render_page.py \u003Curl> --mode auto` (SPA-aware SSRF protection via `url_safety`)\n- No credentials stored or transmitted\n- Output files contain no PII or API keys\n- DataForSEO cost checks run before every API call\n\n## FLOW Framework Integration\n\nFor prompt-guided keyword research and gap analysis, use `\u002Fseo flow find [url|topic]`: FLOW's 5 find-stage prompts complement the SERP-overlap clustering methodology with structured discovery prompts.\n","# Semantic Topic Clustering\n\nSERP-overlap-driven keyword clustering for content architecture. Groups keywords\nby how Google actually ranks them (shared top-10 results), not by text similarity.\nDesigns hub-and-spoke content clusters with internal link matrices and generates\ninteractive cluster map visualizations.\n\n**Scripts:** Located at the plugin root `scripts\u002F` directory.\n\n---\n\n## Quick Reference\n\n| Command | What it does |\n|---------|-------------|\n| `\u002Fseo cluster plan \u003Cseed-keyword>` | Full planning workflow: expand, cluster, architect, visualize |\n| `\u002Fseo cluster plan --from strategy` | Import from existing `\u002Fseo plan` output |\n| `\u002Fseo cluster execute` | Execute plan: create content via claude-blog or output briefs |\n| `\u002Fseo cluster map` | Regenerate the interactive cluster visualization |\n\n---\n\n## Planning Workflow\n\n### Step 1: Seed Keyword Expansion\n\nExpand the seed keyword into 30-50 variants using WebSearch:\n\n1. **Related searches**: Search the seed, extract \"related searches\" and \"people also search for\"\n2. **People Also Ask (PAA)**: Extract all PAA questions from SERP results\n3. **Long-tail modifiers**: Append common modifiers: \"best\", \"how to\", \"vs\", \"for beginners\", \"tools\", \"examples\", \"guide\", \"template\", \"mistakes\", \"checklist\"\n4. **Question mining**: Generate who\u002Fwhat\u002Fwhen\u002Fwhere\u002Fwhy\u002Fhow variants\n5. **Intent modifiers**: Add commercial modifiers: \"pricing\", \"review\", \"alternative\", \"comparison\", \"free\", \"top\"\n\n**Deduplication:** Normalize variants (lowercase, strip articles), remove exact duplicates.\nTarget: 30-50 unique keyword variants. If under 30, run a second expansion pass\nwith the top PAA questions as seeds.\n\n### Step 2: SERP Overlap Clustering\n\nThis is the core differentiator. Load `references\u002Fserp-overlap-methodology.md` for\nthe full algorithm.\n\n**Process:**\n1. Group keywords by initial intent guess (reduces pairwise comparisons)\n2. For each candidate pair within a group, WebSearch both keywords\n3. Count shared URLs in the top 10 organic results (ignore ads, featured snippets, PAA)\n4. Apply thresholds:\n\n| Shared Results | Relationship | Action |\n|---------------|-------------|--------|\n| 7-10 | Same post | Merge into single target page |\n| 4-6 | Same cluster | Group under same spoke cluster |\n| 2-3 | Interlink | Place in adjacent clusters, add cross-links |\n| 0-1 | Separate | Assign to different clusters or exclude |\n\n**Optimization:** With 40 keywords, full pairwise = 780 comparisons. Instead:\n- Pre-group by intent (4 groups of ~10 = 4 x 45 = 180 comparisons)\n- Only cross-check group boundary keywords\n- Skip pairs where both are long-tail variants of the same head term (assume same cluster)\n\n**DataForSEO integration:** If DataForSEO MCP is available, use `serp_organic_live_advanced`\ninstead of WebSearch for SERP data. Run `\u002Fhome\u002Fnoirelab\u002F.config\u002Fopencode\u002Fskills\u002Fseo\u002Fbin\u002Fclaude-seo run dataforseo_costs.py check serp_organic_live_advanced --count N`\nbefore each batch. If `\"status\": \"needs_approval\"`, show cost estimate and ask user.\nIf `\"status\": \"blocked\"`, fall back to WebSearch.\n\n### Step 3: Intent Classification\n\nClassify each keyword into one of four intent categories:\n\n| Intent | Signals | Include in Clusters? |\n|--------|---------|---------------------|\n| Informational | how, what, why, guide, tutorial, learn | Yes |\n| Commercial | best, top, review, comparison, vs, alternative | Yes |\n| Transactional | buy, price, discount, coupon, order, sign up | Yes |\n| Navigational | brand names, specific product names, login | No (exclude) |\n\nRemove navigational keywords from clustering. Flag borderline cases for\nmanual review. Keywords can have mixed intent (e.g., \"best CRM software\" is\nboth commercial and informational) -- classify by dominant intent.\n\n### Step 4: Hub-and-Spoke Architecture\n\nLoad `references\u002Fhub-spoke-architecture.md` for full specifications.\n\n**Design the cluster structure:**\n\n1. **Select the pillar keyword**: Highest volume, broadest intent, most SERP overlap with other keywords\n2. **Group spokes into clusters**: Each cluster is a subtopic area (2-5 clusters per pillar)\n3. **Assign posts to clusters**: Each cluster gets 2-4 spoke posts\n4. **Select templates per post**: Based on intent classification:\n\n| Intent Pattern | Template Options |\n|---------------|-----------------|\n| Informational (broad) | ultimate-guide |\n| Informational (how) | how-to |\n| Informational (list) | listicle |\n| Informational (concept) | explainer |\n| Commercial (compare) | comparison |\n| Commercial (evaluate) | review |\n| Commercial (rank) | best-of |\n| Transactional | landing-page |\n\n5. **Set word count targets:**\n   - Pillar page: 2500-4000 words\n   - Spoke posts: 1200-1800 words\n\n6. **Cannibalization check**: No two posts share the same primary keyword. If SERP\n   overlap is 7+, merge those keywords into a single post targeting both.\n\n### Step 5: Internal Link Matrix\n\nDesign the bidirectional linking structure:\n\n| Link Type | Direction | Requirement |\n|-----------|-----------|-------------|\n| Spoke to pillar | spoke -> pillar | Mandatory (every spoke) |\n| Pillar to spoke | pillar -> spoke | Mandatory (every spoke) |\n| Spoke to spoke (within cluster) | spoke \u003C-> spoke | 2-3 links per post |\n| Cross-cluster | spoke -> spoke (other cluster) | 0-1 links per post |\n\n**Rules:**\n- Every post must have minimum 3 incoming internal links\n- No orphan pages (every post reachable from pillar in 2 clicks)\n- Anchor text must use target keyword or close variant (no \"click here\")\n- Link placement: within body content, not just navigation\u002Fsidebar\n\nGenerate the link matrix as a JSON adjacency list:\n```json\n{\n  \"links\": [\n    { \"from\": \"pillar\", \"to\": \"cluster-0-post-0\", \"type\": \"mandatory\", \"anchor\": \"keyword\" },\n    { \"from\": \"cluster-0-post-0\", \"to\": \"pillar\", \"type\": \"mandatory\", \"anchor\": \"keyword\" }\n  ]\n}\n```\n\n### Step 6: Interactive Cluster Map\n\nGenerate `cluster-map.html` using the template at `templates\u002Fcluster-map.html`.\n\n1. Read the template file\n2. Build the `CLUSTER_DATA` JSON object from the cluster plan:\n   ```javascript\n   {\n     pillar: { title, keyword, volume, template, wordCount, url },\n     clusters: [{ name, color, posts: [{ title, keyword, volume, template, wordCount, url, status }] }],\n     links: [{ from, to, type }],\n     meta: { totalPosts, totalClusters, totalLinks, estimatedWords }\n   }\n   ```\n3. Replace the `CLUSTER_DATA` placeholder in the template with the actual JSON\n4. Write the completed HTML file to the output directory\n5. Inform user: \"Open `cluster-map.html` in a browser to explore the interactive cluster map.\"\n\n---\n\n## Strategy Import\n\nWhen invoked with `--from strategy`:\n\n1. Look for the most recent `\u002Fseo plan` output in the current directory (search for\n   files matching `*SEO*Plan*`, `*strategy*`, `*content-strategy*`)\n2. Parse markdown tables for: keywords, page types, content pillars, URL structures\n3. Validate extracted data: check for duplicates, missing keywords, incomplete entries\n4. Enrich with SERP data: run SERP overlap analysis on extracted keywords\n5. Build cluster plan using the imported keywords as the starting set (skip Step 1)\n\nIf no strategy file is found, prompt the user: \"No existing SEO plan found in the\ncurrent directory. Run `\u002Fseo plan` first, or provide a seed keyword for fresh clustering.\"\n\n---\n\n## Execution Workflow\n\nWhen `\u002Fseo cluster execute` is invoked:\n\n### Check for claude-blog\n\n```\nTest: Does ~\u002F.claude\u002Fskills\u002Fblog\u002FSKILL.md exist?\n```\n\n**If claude-blog IS installed:**\n\n1. Load `references\u002Fexecution-workflow.md` for the full algorithm\n2. Read `cluster-plan.json` from the current directory\n3. Check for resume state: scan output directory for already-written posts\n4. Execute in priority order: pillar first, then spokes by volume (highest first)\n5. For each post, invoke the `blog-write` skill with cluster context:\n   - Cluster role (pillar or spoke)\n   - Position in cluster (cluster index, post index)\n   - Target keyword and secondary keywords\n   - Template type and word count target\n   - Internal links to include (with anchors)\n   - Links to receive from future posts (placeholder markers)\n6. After each post is written, scan previous posts for backward link placeholders\n   and inject the new post's URL\n7. After all posts are written, generate the cluster scorecard\n\n**If claude-blog is NOT installed:**\n\n1. Generate detailed content briefs for each post in the cluster plan\n2. Each brief includes:\n   - Title and meta description\n   - Primary keyword and secondary keywords\n   - Template type and suggested structure (H2\u002FH3 outline)\n   - Word count target\n   - Internal links to include (with anchor text)\n   - Key points to cover\n   - Competing pages to differentiate from\n3. Write briefs to `cluster-briefs\u002F` directory as individual markdown files\n4. Inform user: \"Install [claude-blog](https:\u002F\u002Fgithub.com\u002FAgriciDaniel\u002Fclaude-blog)\n   to auto-create content. Briefs saved to `cluster-briefs\u002F`.\"\n\n---\n\n## Cluster Scorecard\n\nPost-execution quality report. Run automatically after `\u002Fseo cluster execute` or\non demand via analysis of the output directory.\n\n| Metric | Target | How Measured |\n|--------|--------|-------------|\n| Coverage | 100% | Posts written \u002F posts planned |\n| Link Density | 3+ per post | Count internal links per post |\n| Orphan Pages | 0 | Posts with \u003C 1 incoming link |\n| Cannibalization | 0 conflicts | Check for duplicate primary keywords |\n| Image Count | 1+ per post | Posts with at least one image |\n| Pillar Links | 100% | All spokes link to pillar and vice versa |\n| Cross-Links | 80%+ | Recommended spoke-to-spoke links implemented |\n| Content Gaps | 0 | Planned posts that were skipped or incomplete |\n\n---\n\n## Map Regeneration\n\nWhen `\u002Fseo cluster map` is invoked:\n\n1. Read `cluster-plan.json` from the current directory\n2. Scan output directory and update post statuses (planned vs written)\n3. Regenerate `cluster-map.html` with updated statuses\n4. Report: posts written vs planned, link completion percentage\n\n---\n\n## Output Files\n\nAll outputs are written to the current working directory:\n\n| File | Description |\n|------|-------------|\n| `cluster-plan.json` | Machine-readable cluster plan (full data) |\n| `cluster-plan.md` | Human-readable cluster plan summary |\n| `cluster-map.html` | Interactive SVG visualization |\n| `cluster-briefs\u002F` | Content briefs (if no claude-blog) |\n| `cluster-scorecard.md` | Post-execution quality report |\n\n---\n\n## Cross-Skill Integration\n\n| Skill | Relationship |\n|-------|-------------|\n| `seo-plan` | Import source: strategy import reads seo-plan output |\n| `seo-content` | Quality check: E-E-A-T validation of generated content |\n| `seo-schema` | Schema markup: Article, BreadcrumbList, ItemList for cluster pages |\n| `seo-dataforseo` | Data source: SERP data when DataForSEO MCP is available |\n| `seo-google` | Reporting: generate PDF report of cluster plan and scorecard |\n\nAfter cluster planning or execution completes, offer:\n\"Generate a PDF report? Use `\u002Fseo google report`\"\n\n---\n\n## Error Handling\n\n| Error | Cause | Resolution |\n|-------|-------|------------|\n| \"No seed keyword provided\" | Missing argument | Prompt user for seed keyword or URL |\n| \"Insufficient keyword variants\" | Expansion yielded \u003C 15 keywords | Run second expansion pass with PAA questions |\n| \"SERP data unavailable\" | WebSearch and DataForSEO both failing | Retry after 30s; if persistent, use intent-only clustering with warning |\n| \"No strategy file found\" | `--from strategy` but no plan exists | Prompt user to run `\u002Fseo plan` first |\n| \"cluster-plan.json not found\" | Execute without planning | Prompt user to run `\u002Fseo cluster plan` first |\n| \"claude-blog not installed\" | Execute attempted without blog skill | Generate content briefs instead; suggest installation |\n| \"DataForSEO budget exceeded\" | Cost check returned \"blocked\" | Fall back to WebSearch; inform user |\n| \"Duplicate primary keywords\" | Cannibalization detected | Merge affected posts or reassign keywords |\n| \"Orphan page detected\" | Post missing incoming links | Add links from nearest cluster siblings |\n| \"Resume state corrupted\" | Mismatch between plan and output | Rebuild state from output directory scan |\n\n---\n\n## Security\n\n- All URLs fetched via `\u002Fhome\u002Fnoirelab\u002F.config\u002Fopencode\u002Fskills\u002Fseo\u002Fbin\u002Fclaude-seo run render_page.py \u003Curl> --mode auto` (SPA-aware SSRF protection via `url_safety`)\n- No credentials stored or transmitted\n- Output files contain no PII or API keys\n- DataForSEO cost checks run before every API call\n\n## FLOW Framework Integration\n\nFor prompt-guided keyword research and gap analysis, use `\u002Fseo flow find [url|topic]`: FLOW's 5 find-stage prompts complement the SERP-overlap clustering methodology with structured discovery prompts.\n",{"name":5,"license":15,"metadata":66,"description":70,"argument-hint":71,"user-invocable":72},{"author":67,"version":68,"category":35,"original_author":69},"AgriciDaniel","2.2.4","Lutfiya Miller (Pro Hub Challenge Winner)","SERP-based semantic topic clustering for content architecture planning. Groups keywords by actual Google SERP overlap (not text similarity), designs hub-and-spoke content clusters with internal link matrices, and generates interactive visualizations. Optionally executes content creation if claude-blog is installed. Use when user says \"topic cluster\", \"content cluster\", \"semantic clustering\", \"pillar page\", \"hub and spoke\", \"content architecture\", \"keyword grouping\", or \"cluster plan\".\n","\u003Cseed-keyword or url>",true,"main","d8378e461d60315bd1ca8a6e71716df39b6abe4a","configs\u002Fcommon\u002Fclaude\u002Fskills\u002Fseo-cluster","https:\u002F\u002Fgithub.com\u002Fnoirelab\u002Flinux-setup","https:\u002F\u002Fgithub.com\u002Fnoirelab","Artificial Intelligence Researcher |\r\nComputer Science majoring at UTFPR","2026-11-11 09:12:51",{"source":8,"skillName":5,"commands":81,"prompt":91,"zipAvailable":72},[82,85,88],{"tool":83,"command":84},"npx","npx skills add noirelab\u002Flinux-setup --skill seo-cluster",{"tool":86,"command":87},"bunx","bunx skills add noirelab\u002Flinux-setup --skill seo-cluster",{"tool":89,"command":90},"pnpm","pnpm dlx skills add noirelab\u002Flinux-setup --skill seo-cluster","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fnoirelab\u002Flinux-setup\u002Fblob\u002Fmain\u002Fconfigs\u002Fcommon\u002Fclaude\u002Fskills\u002Fseo-cluster\u002FSKILL.md",[93,99,103,107,111],{"path":94,"resourceType":95,"resourceTypeDesc":96,"sizeBytes":97,"sourceHtmlUrl":98},"NOTICE.md","REFERENCE","参考文档",1198,"https:\u002F\u002Fgithub.com\u002Fknacklabs\u002Fsymphony-forge\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ftest-audit\u002FNOTICE.md",{"path":100,"resourceType":95,"resourceTypeDesc":96,"sizeBytes":101,"sourceHtmlUrl":102},"references\u002Fexecution-workflow.md",6455,"https:\u002F\u002Fgithub.com\u002Fnoirelab\u002Flinux-setup\u002Fblob\u002Fmain\u002Fconfigs\u002Fcommon\u002Fclaude\u002Fskills\u002Fseo-cluster\u002Freferences\u002Fexecution-workflow.md",{"path":104,"resourceType":95,"resourceTypeDesc":96,"sizeBytes":105,"sourceHtmlUrl":106},"references\u002Fhub-spoke-architecture.md",6291,"https:\u002F\u002Fgithub.com\u002Fnoirelab\u002Flinux-setup\u002Fblob\u002Fmain\u002Fconfigs\u002Fcommon\u002Fclaude\u002Fskills\u002Fseo-cluster\u002Freferences\u002Fhub-spoke-architecture.md",{"path":108,"resourceType":95,"resourceTypeDesc":96,"sizeBytes":109,"sourceHtmlUrl":110},"references\u002Fserp-overlap-methodology.md",4514,"https:\u002F\u002Fgithub.com\u002Fnoirelab\u002Flinux-setup\u002Fblob\u002Fmain\u002Fconfigs\u002Fcommon\u002Fclaude\u002Fskills\u002Fseo-cluster\u002Freferences\u002Fserp-overlap-methodology.md",{"path":112,"resourceType":95,"resourceTypeDesc":96,"sizeBytes":113,"sourceHtmlUrl":114},"templates\u002Fcluster-map.html",18882,"https:\u002F\u002Fgithub.com\u002Fnoirelab\u002Flinux-setup\u002Fblob\u002Fmain\u002Fconfigs\u002Fcommon\u002Fclaude\u002Fskills\u002Fseo-cluster\u002Ftemplates\u002Fcluster-map.html",[116,147,182,208,228,244],{"skillNo":117,"name":118,"nameZh":6,"description":119,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":120,"licenseFileUrl":121,"sourceHtmlUrl":122,"categories":123,"keywords":127,"publishedAt":61},"01M6C2Q1S6HCS7Y9TBQ2PJ2NAM","skill-creator","Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fnoirelab\u002Flinux-setup\u002Fblob\u002Fmain\u002Fconfigs\u002Fcommon\u002Fclaude\u002Fskills\u002Fsynced\u002Ffcfa9261-0073-49cb-8312-19eea68f0c58_391f3767-a35d-4280-81a1-363cd861a8e6\u002Fskill-creator\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fnoirelab\u002Flinux-setup\u002Fblob\u002Fmain\u002Fconfigs\u002Fcommon\u002Fclaude\u002Fskills\u002Fsynced\u002Ffcfa9261-0073-49cb-8312-19eea68f0c58_391f3767-a35d-4280-81a1-363cd861a8e6\u002Fskill-creator\u002FSKILL.md",[124],{"categoryNo":125,"categoryName":126,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[128,129,130,131,132,133,134,135,136,137,138,139,47,140,141,142,143,144,145,40,146],"skill creator","creator","performance","run evals","test","variance analysis","create","run","benchmark","analysis","improve","edit","evals","optimize","existing skill","for","accuracy","scratch","new",{"skillNo":148,"name":149,"nameZh":6,"description":150,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":120,"licenseFileUrl":6,"sourceHtmlUrl":151,"categories":152,"keywords":153,"publishedAt":61},"01M6C2ZRGJ9HN4N9BP24DPNCR4","returns-reverse-logistics","Codified expertise for returns authorization, receipt and inspection, disposition decisions, refund processing, fraud detection, and warranty claims management. Informed by returns operations managers with 15+ years experience. Includes grading frameworks, disposition economics, fraud pattern recognition, and vendor recovery processes. Use when handling product returns, reverse logistics, refund decisions, return fraud detection, or warranty claims.","https:\u002F\u002Fgithub.com\u002Fnoirelab\u002Flinux-setup\u002Fblob\u002Fmain\u002Fconfigs\u002Fcommon\u002Fopencode_home\u002Fskills\u002Freturns-reverse-logistics\u002FSKILL.md",[],[154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,47,170,171,172,173,174,175,176,177,178,143,179,180,40,181],"returns","returns reverse logistics","reverse logistics","reverse","github","code","product","https","authorization","detection","fraud","recovery","decisions","management","operations","vendor","claims management","frameworks","fraud detection","pattern","grading","processing","economics","managers","claims","expertise","refund","inspection",{"skillNo":183,"name":184,"nameZh":6,"description":185,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":120,"licenseFileUrl":6,"sourceHtmlUrl":186,"categories":187,"keywords":188,"publishedAt":61},"01M6C2ZGWSNZGZZQ5VCFWHD2QT","quality-nonconformance","Codified expertise for quality control, non-conformance investigation, root cause analysis, corrective action, and supplier quality management in regulated manufacturing. Informed by quality engineers with 15+ years experience across FDA, IATF 16949, and AS9100 environments. Includes NCR lifecycle management, CAPA systems, SPC interpretation, and audit methodology. Use when investigating non-conformances, performing root cause analysis, managing CAPAs, interpreting SPC data, or handling supplier quality issues.","https:\u002F\u002Fgithub.com\u002Fnoirelab\u002Flinux-setup\u002Fblob\u002Fmain\u002Fconfigs\u002Fcommon\u002Fopencode_home\u002Fskills\u002Fquality-nonconformance\u002FSKILL.md",[],[189,190,158,191,192,159,193,161,194,195,196,197,137,198,167,47,199,200,201,202,203,204,205,206,143,179,207,40],"quality","quality nonconformance","root cause analysis","interpretation","issues","audit","data","methodology","lifecycle","investigation","fda","across","manufacturing","environments","corrective action","supplier quality management","action","conformance","root cause",{"skillNo":209,"name":210,"nameZh":6,"description":211,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":120,"licenseFileUrl":6,"sourceHtmlUrl":212,"categories":213,"keywords":214,"publishedAt":61},"01M6C2Z8VQ9AYZYKPZW1KD9DX5","production-scheduling","Codified expertise for production scheduling, job sequencing, line balancing, changeover optimization, and bottleneck resolution in discrete and batch manufacturing. Informed by production schedulers with 15+ years experience. Includes TOC\u002Fdrum-buffer-rope, SMED, OEE analysis, disruption response frameworks, and ERP\u002FMES interaction patterns. Use when scheduling production, resolving bottlenecks, optimizing changeovers, responding to disruptions, or balancing manufacturing lines.","https:\u002F\u002Fgithub.com\u002Fnoirelab\u002Flinux-setup\u002Fblob\u002Fmain\u002Fconfigs\u002Fcommon\u002Fopencode_home\u002Fskills\u002Fproduction-scheduling\u002FSKILL.md",[],[215,216,217,158,159,218,161,219,137,220,221,47,222,223,171,224,225,226,201,227,143,179,40],"Production","production scheduling","scheduling","batch","Optimization","erp","bottleneck","toc","patterns","line","interaction","sequencing","buffer",{"skillNo":229,"name":230,"nameZh":6,"description":231,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":120,"licenseFileUrl":6,"sourceHtmlUrl":232,"categories":233,"keywords":234,"publishedAt":61},"01M6C2YR01NNN2Z1G18RN7F9MP","logistics-exception-management","Codified expertise for handling freight exceptions, shipment delays, damages, losses, and carrier disputes. Informed by logistics professionals with 15+ years operational experience. Includes escalation protocols, carrier-specific behaviors, claims procedures, and judgment frameworks. Use when handling shipping exceptions, freight claims, delivery issues, or carrier disputes.","https:\u002F\u002Fgithub.com\u002Fnoirelab\u002Flinux-setup\u002Fblob\u002Fmain\u002Fconfigs\u002Fcommon\u002Fopencode_home\u002Fskills\u002Flogistics-exception-management\u002FSKILL.md",[],[167,235,158,159,236,193,237,161,238,47,171,239,240,241,242,178,143,179,243,40],"logistics exception management","shipping","protocols","escalation","delivery","claude-code","judgment","with","freight",{"skillNo":245,"name":246,"nameZh":6,"description":247,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":120,"licenseFileUrl":6,"sourceHtmlUrl":248,"categories":249,"keywords":250,"publishedAt":61},"01M6C2YC5PXTEJYBJJQZVXPQS0","inventory-demand-planning","Codified expertise for demand forecasting, safety stock optimization, replenishment planning, and promotional lift estimation at multi-location retailers. Informed by demand planners with 15+ years experience managing hundreds of SKUs. Includes forecasting method selection, ABC\u002FXYZ analysis, seasonal transition management, and vendor negotiation frameworks. Use when forecasting demand, setting safety stock, planning replenishment, managing promotions, or optimizing inventory levels.","https:\u002F\u002Fgithub.com\u002Fnoirelab\u002Flinux-setup\u002Fblob\u002Fmain\u002Fconfigs\u002Fcommon\u002Fopencode_home\u002Fskills\u002Finventory-demand-planning\u002FSKILL.md",[],[59,251,252,253,158,159,161,254,219,255,137,167,169,256,47,257,171,258,259,260,261,262,143,179,263,40],"inventory","inventory demand planning","demand planning","negotiation","Safety","forecasting","selection","estimation","stock","seasonal","lift","at","transition"]