{"id":"c9e92a89-3590-423b-9c69-11ee04e72107","shortId":"PznjB5","kind":"skill","title":"blog-cannibalization","tagline":"Detect keyword cannibalization across blog posts by extracting primary keywords from titles and headings, clustering semantically similar targets, and flagging posts competing for the same search intent. Supports local-only mode (grep-based) and DataForSEO API mode (Page Intersec","description":"# Blog Cannibalization - Keyword Overlap Detection\n\nDetect when multiple blog posts compete for the same search keywords. Two modes:\nlocal-only analysis (default) and DataForSEO API mode for SERP-level data.\n\n## Two Modes\n\n| Mode | Flag | Cost | Data Source |\n|------|------|------|-------------|\n| Local | (default) | Free | File content analysis via Grep/Read |\n| API | `--api` | ~$0.01/call | DataForSEO Page Intersection + Ranked Keywords |\n\nLocal mode works without any API keys. API mode requires DataForSEO credentials\nset as environment variables: `DATAFORSEO_LOGIN` and `DATAFORSEO_PASSWORD`.\n\n## Local Mode Workflow\n\n### Step 1: Scan Blog Files\n\nUse Glob to find all content files in the target directory:\n- Patterns: `**/*.md`, `**/*.mdx`, `**/*.html`\n- Skip files in `node_modules/`, `.git/`, `drafts/`\n\n### Step 2: Extract Primary Keywords\n\nFor each file, read and extract keyword signals from:\n- **Title tag** or H1 heading (highest weight)\n- **H2 headings** (medium weight)\n- **First paragraph** (supporting signal)\n- **Meta description** if present in frontmatter\n\nPrimary keyword extraction method:\n1. Tokenize title and H1 into 1-gram, 2-gram, and 3-gram phrases\n2. Score each phrase by frequency across title + H2s + first paragraph\n3. Select the top-scoring 2-3 word phrase as the primary keyword\n4. Record secondary keywords from H2 headings\n\n### Step 3: Cluster by Similarity\n\nGroup posts into clusters using these matching rules (in priority order):\n\n1. **Exact match** - identical primary keyword across 2+ posts\n2. **Stem match** - same root word (e.g., \"optimize\" vs \"optimization\")\n3. **Semantic overlap** - Claude determines that two keywords target the same\n   search intent (e.g., \"best CRM software\" vs \"top CRM tools 2026\")\n4. **Subset match** - one keyword contains another (e.g., \"email marketing\"\n   vs \"email marketing for startups\")\n\n### Step 4: Score and Flag\n\nFor each cluster with 2+ posts, assess severity and generate a recommendation.\n\n### Step 5: Output Report\n\nDisplay the results table and per-cluster recommendations.\n\n## API Mode Workflow (DataForSEO)\n\nRequires the `--api` flag. Uses WebFetch to call DataForSEO endpoints.\n\n### Endpoints Used\n\n**Page Intersection** - find keywords where multiple URLs rank:\n```\nPOST https://api.dataforseo.com/v3/dataforseo_labs/google/page_intersection/live\nAuthorization: Basic <base64(login:password)>\n\n{\n  \"pages\": {\n    \"1\": \"https://example.com/post-a\",\n    \"2\": \"https://example.com/post-b\"\n  },\n  \"language_code\": \"en\",\n  \"location_code\": 2840\n}\n```\nCost: ~$0.01 per call. Returns overlapping keywords with position, volume, CPC.\n\n**Ranked Keywords** - get all keywords a single URL ranks for:\n```\nPOST https://api.dataforseo.com/v3/dataforseo_labs/google/ranked_keywords/live\n\n{\n  \"target\": \"https://example.com/post-a\",\n  \"language_code\": \"en\",\n  \"location_code\": 2840\n}\n```\n\n### API Analysis Steps\n\n1. Collect all published URLs from the user (or sitemap)\n2. Run Ranked Keywords for each URL to build keyword profiles\n3. Run Page Intersection for URL pairs that share keyword clusters\n4. Calculate severity using the formula below\n5. Output enriched report with search volume and position data\n\n## Severity Scoring\n\nFour severity levels based on overlap signals:\n\n| Level | Criteria | Action Urgency |\n|-------|----------|----------------|\n| Critical | Same exact keyword, both pages in top 20 | Immediate |\n| High | Same keyword cluster, one page outranks the other | This week |\n| Medium | Related keywords with partial SERP overlap | This month |\n| Low | Semantic similarity but different confirmed intents | Monitor |\n\n### Severity Formula (API Mode)\n\n```\nseverity_score = overlap_count x avg_search_volume x (1 / position_gap)\n```\n\nWhere:\n- `overlap_count` = number of shared ranking keywords\n- `avg_search_volume` = mean monthly volume of shared keywords\n- `position_gap` = absolute difference in average ranking position (min 1)\n\nHigher score = more urgent cannibalization problem.\n\n### Severity Heuristic (Local Mode)\n\nWithout SERP data, use a simplified scoring:\n- **Critical**: Exact primary keyword match between posts\n- **High**: Stem match on primary keyword, or 3+ shared H2 keywords\n- **Medium**: Semantic overlap on primary keyword\n- **Low**: Subset match only, or shared secondary keywords\n\n## Output Format\n\n### Summary Table\n\n```\n| Post A | Post B | Shared Keywords | Severity | Recommendation |\n|--------|--------|-----------------|----------|----------------|\n| /best-crm-tools | /top-crm-software | best crm, crm tools, crm software | Critical | MERGE |\n| /email-tips | /email-marketing-guide | email marketing | High | DIFFERENTIATE |\n| /seo-basics | /seo-for-beginners | seo basics, beginner seo | Critical | CANONICAL |\n| /react-hooks | /react-state-mgmt | react, state | Low | NO ACTION |\n```\n\n### Per-Cluster Detail\n\nFor each flagged cluster, provide:\n- Both post titles and URLs\n- Full list of overlapping keywords (with volume if API mode)\n- Which post is stronger (more comprehensive, better structured)\n- Specific recommendation with rationale\n\n## Recommendations\n\nFour possible actions for each cannibalization cluster:\n\n### MERGE\nWhen both pages are thin or cover the same intent with similar depth.\n- Combine the best content from both into one comprehensive post\n- 301 redirect the weaker URL to the merged post\n- Preserve all internal links pointing to either URL\n\n### DIFFERENTIATE\nWhen pages serve different intents but keyword targeting overlaps.\n- Shift the primary keyword of the weaker post to a related long-tail\n- Update the title, H1, and meta description to reflect the new focus\n- Add internal links between the two posts to signal distinct topics\n\n### CANONICAL\nWhen one post is clearly the authority and the other is a lesser duplicate.\n- Add `rel=\"canonical\"` on the weaker page pointing to the authority\n- Consider noindexing the weaker page if it adds no unique value\n- Link from the weaker page to the authority page\n\n### NO ACTION\nWhen intent is genuinely different despite surface-level keyword similarity.\n- Document the reasoning for future audits\n- Monitor rankings quarterly for any position changes\n- Re-evaluate if either post drops in rankings\n\n## Error Handling\n\n- **No blog files found**: If the directory contains no .md, .mdx, or .html files, report \"No blog files found in [directory]\" and suggest checking the path\n- **DataForSEO credentials missing**: In API mode, if credentials are not configured, fall back to local mode automatically and notify the user\n- **API rate limits**: DataForSEO has per-minute rate limits. If a 429 response is received, wait and retry once. If it persists, switch to local mode for remaining URLs\n- **WebFetch failures**: If a source URL is unreachable, skip it and note \"Unable to verify - source unavailable\" in the report\n- **Single-post directory**: If only one blog post exists, report \"Cannibalization analysis requires at least 2 posts\" and exit gracefully","tags":["blog","cannibalization","claude","agricidaniel","agent-skills","ai-citations","ai-content","ai-marketing","ai-marketing-hub","blog-writing","claude-code","claude-code-skill"],"capabilities":["skill","source-agricidaniel","skill-blog-cannibalization","topic-agent-skills","topic-ai-citations","topic-ai-content","topic-ai-marketing","topic-ai-marketing-hub","topic-blog","topic-blog-writing","topic-claude-code","topic-claude-code-skill","topic-claude-plugin","topic-claude-skill","topic-content-creation"],"categories":["claude-blog"],"synonyms":[],"warnings":[],"endpointUrl":"https://skills.sh/AgriciDaniel/claude-blog/blog-cannibalization","protocol":"skill","transport":"skills-sh","auth":{"type":"none","details":{"cli":"npx skills add AgriciDaniel/claude-blog","source_repo":"https://github.com/AgriciDaniel/claude-blog","install_from":"skills.sh"}},"qualityScore":"0.700","qualityRationale":"deterministic score 0.70 from registry signals: · indexed on github topic:agent-skills · 753 github stars · SKILL.md body (6,957 chars)","verified":false,"liveness":"unknown","lastLivenessCheck":null,"agentReviews":{"count":0,"score_avg":null,"cost_usd_avg":null,"success_rate":null,"latency_p50_ms":null,"narrative_summary":null,"summary_updated_at":null},"enrichmentModel":"deterministic:skill-github:v1","enrichmentVersion":1,"enrichedAt":"2026-05-18T18:53:29.541Z","embedding":null,"createdAt":"2026-05-18T12:54:52.009Z","updatedAt":"2026-05-18T18:53:29.541Z","lastSeenAt":"2026-05-18T18:53:29.541Z","tsv":"'-3':223 '/best-crm-tools':629 '/call':95 '/email-marketing-guide':640 '/email-tips':639 '/post-a':376,415 '/post-b':380 '/react-hooks':653 '/react-state-mgmt':654 '/seo-basics':645 '/seo-for-beginners':646 '/top-crm-software':630 '/v3/dataforseo_labs/google/page_intersection/live':366 '/v3/dataforseo_labs/google/ranked_keywords/live':411 '0.01':94,388 '1':126,191,197,253,373,425,538,567 '2':153,199,205,222,260,262,318,377,435,988 '20':495 '2026':293 '2840':386,421 '3':202,216,238,272,446,599 '301':728 '4':230,294,310,457 '429':934 '5':327,464 'absolut':560 'across':7,211,259 'action':485,659,699,839 'add':781,807,825 'analysi':66,89,423,984 'anoth':300 'api':41,70,92,93,106,108,339,345,422,527,682,905,922 'api.dataforseo.com':365,410 'api.dataforseo.com/v3/dataforseo_labs/google/page_intersection/live':364 'api.dataforseo.com/v3/dataforseo_labs/google/ranked_keywords/live':409 'assess':320 'audit':856 'author':367,799,817,836 'automat':917 'averag':563 'avg':534,549 'b':624 'back':913 'base':38,479 'base64':369 'basic':368,648 'beginn':649 'best':286,631,720 'better':690 'blog':2,8,45,53,128,876,891,979 'blog-cannib':1 'build':443 'calcul':458 'call':350,390 'cannib':3,6,46,572,702,983 'canon':652,792,809 'chang':863 'check':898 'claud':275 'clear':797 'cluster':18,239,245,316,337,456,500,662,667,703 'code':382,385,417,420 'collect':426 'combin':718 'compet':25,55 'comprehens':689,726 'configur':911 'confirm':522 'consid':818 'contain':299,882 'content':88,135,721 'cost':81,387 'count':532,543 'cover':711 'cpc':397 'credenti':112,902,908 'criteria':484 'critic':487,585,637,651 'crm':287,291,632,633,635 'data':76,82,473,580 'dataforseo':40,69,96,111,117,120,342,351,901,925 'default':67,85 'depth':717 'descript':182,775 'despit':845 'detail':663 'detect':4,49,50 'determin':276 'differ':521,561,749,844 'differenti':644,745 'directori':140,881,895,975 'display':330 'distinct':790 'document':851 'draft':151 'drop':870 'duplic':806 'e.g':268,285,301 'either':743,868 'email':302,305,641 'en':383,418 'endpoint':352,353 'enrich':466 'environ':115 'error':873 'evalu':866 'exact':254,489,586 'example.com':375,379,414 'example.com/post-a':374,413 'example.com/post-b':378 'exist':981 'exit':991 'extract':11,154,162,189 'failur':953 'fall':912 'file':87,129,136,146,159,877,888,892 'find':133,357 'first':177,214 'flag':23,80,313,346,666 'focus':780 'format':618 'formula':462,526 'found':878,893 'four':476,697 'free':86 'frequenc':210 'frontmatt':186 'full':674 'futur':855 'gap':540,559 'generat':323 'genuin':843 'get':400 'git':150 'glob':131 'grace':992 'gram':198,200,203 'grep':37 'grep-bas':36 'grep/read':91 'group':242 'h1':169,195,772 'h2':173,235,601 'h2s':213 'handl':874 'head':17,170,174,236 'heurist':575 'high':497,592,643 'higher':568 'highest':171 'html':144,887 'ident':256 'immedi':496 'intent':30,284,523,714,750,841 'intern':739,782 'intersec':44 'intersect':98,356,449 'key':107 'keyword':5,13,47,60,100,156,163,188,229,233,258,279,298,358,393,399,402,438,444,455,490,499,510,548,557,588,597,602,608,616,626,678,752,758,849 'languag':381,416 'least':987 'lesser':805 'level':75,478,483,848 'limit':924,931 'link':740,783,829 'list':675 'local':33,64,84,101,122,576,915,947 'local-on':32,63 'locat':384,419 'login':118,370 'long':767 'long-tail':766 'low':517,609,657 'market':303,306,642 'match':248,255,264,296,589,594,611 'md':142,884 'mdx':143,885 'mean':552 'medium':175,508,603 'merg':638,704,735 'meta':181,774 'method':190 'min':566 'minut':929 'miss':903 'mode':35,42,62,71,78,79,102,109,123,340,528,577,683,906,916,948 'modul':149 'monitor':524,857 'month':516,553 'multipl':52,360 'new':779 'node':148 'noindex':819 'note':963 'notifi':919 'number':544 'one':297,501,725,794,978 'optim':269,271 'order':252 'output':328,465,617 'outrank':503 'overlap':48,274,392,481,514,531,542,605,677,754 'page':43,97,355,372,448,492,502,707,747,813,822,833,837 'pair':452 'paragraph':178,215 'partial':512 'password':121,371 'path':900 'pattern':141 'per':336,389,661,928 'per-clust':335,660 'per-minut':927 'persist':944 'phrase':204,208,225 'point':741,814 'posit':395,472,539,558,565,862 'possibl':698 'post':9,24,54,243,261,319,363,408,591,621,623,670,685,727,736,762,787,795,869,974,980,989 'present':184 'preserv':737 'primari':12,155,187,228,257,587,596,607,757 'prioriti':251 'problem':573 'profil':445 'provid':668 'publish':428 'quarter':859 'rank':99,362,398,406,437,547,564,858,872 'rate':923,930 'rational':695 're':865 're-evalu':864 'react':655 'read':160 'reason':853 'receiv':937 'recommend':325,338,628,693,696 'record':231 'redirect':729 'reflect':777 'rel':808 'relat':509,765 'remain':950 'report':329,467,889,971,982 'requir':110,343,985 'respons':935 'result':332 'retri':940 'return':391 'root':266 'rule':249 'run':436,447 'scan':127 'score':206,221,311,475,530,569,584 'search':29,59,283,469,535,550 'secondari':232,615 'select':217 'semant':19,273,518,604 'seo':647,650 'serp':74,513,579 'serp-level':73 'serv':748 'set':113 'sever':321,459,474,477,525,529,574,627 'share':454,546,556,600,614,625 'shift':755 'signal':164,180,482,789 'similar':20,241,519,716,850 'simplifi':583 'singl':404,973 'single-post':972 'sitemap':434 'skill' 'skill-blog-cannibalization' 'skip':145,960 'softwar':288,636 'sourc':83,956,967 'source-agricidaniel' 'specif':692 'startup':308 'state':656 'stem':263,593 'step':125,152,237,309,326,424 'stronger':687 'structur':691 'subset':295,610 'suggest':897 'summari':619 'support':31,179 'surfac':847 'surface-level':846 'switch':945 'tabl':333,620 'tag':167 'tail':768 'target':21,139,280,412,753 'thin':709 'titl':15,166,193,212,671,771 'token':192 'tool':292,634 'top':220,290,494 'top-scor':219 'topic':791 'topic-agent-skills' 'topic-ai-citations' 'topic-ai-content' 'topic-ai-marketing' 'topic-ai-marketing-hub' 'topic-blog' 'topic-blog-writing' 'topic-claude-code' 'topic-claude-code-skill' 'topic-claude-plugin' 'topic-claude-skill' 'topic-content-creation' 'two':61,77,278,786 'unabl':964 'unavail':968 'uniqu':827 'unreach':959 'updat':769 'urgenc':486 'urgent':571 'url':361,405,429,441,451,673,732,744,951,957 'use':130,246,347,354,460,581 'user':432,921 'valu':828 'variabl':116 'verifi':966 'via':90 'volum':396,470,536,551,554,680 'vs':270,289,304 'wait':938 'weaker':731,761,812,821,832 'webfetch':348,952 'week':507 'weight':172,176 'without':104,578 'word':224,267 'work':103 'workflow':124,341 'x':533,537","prices":[{"id":"8e8f5dbb-b0fe-43b9-b42c-2560fd59017d","listingId":"c9e92a89-3590-423b-9c69-11ee04e72107","amountUsd":"0","unit":"free","nativeCurrency":null,"nativeAmount":null,"chain":null,"payTo":null,"paymentMethod":"skill-free","isPrimary":true,"details":{"org":"AgriciDaniel","category":"claude-blog","install_from":"skills.sh"},"createdAt":"2026-05-18T12:54:52.009Z"}],"sources":[{"listingId":"c9e92a89-3590-423b-9c69-11ee04e72107","source":"github","sourceId":"AgriciDaniel/claude-blog/blog-cannibalization","sourceUrl":"https://github.com/AgriciDaniel/claude-blog/tree/main/skills/blog-cannibalization","isPrimary":false,"firstSeenAt":"2026-05-18T12:54:52.009Z","lastSeenAt":"2026-05-18T18:53:29.541Z"}],"details":{"listingId":"c9e92a89-3590-423b-9c69-11ee04e72107","quickStartSnippet":null,"exampleRequest":null,"exampleResponse":null,"schema":null,"openapiUrl":null,"agentsTxtUrl":null,"citations":[],"useCases":[],"bestFor":[],"notFor":[],"kindDetails":{"org":"AgriciDaniel","slug":"blog-cannibalization","github":{"repo":"AgriciDaniel/claude-blog","stars":753,"topics":["agent-skills","ai","ai-citations","ai-content","ai-marketing","ai-marketing-hub","blog","blog-writing","claude-code","claude-code-skill","claude-plugin","claude-skill","content-creation","content-optimization","content-strategy","eeat","geo","multilingual","open-source","seo"],"license":"mit","html_url":"https://github.com/AgriciDaniel/claude-blog","pushed_at":"2026-05-15T04:45:18Z","description":"Claude Code blog skill suite: 30 sub-skills, 5 agents, 5-gate v1.9.0 Blog Delivery Contract, dual-optimized for Google rankings and AI citations. Active development at AI-Marketing-Hub/claude-blog (AI Marketing Hub Pro community); public releases ship here.","skill_md_sha":"9af168c50547dd0e8a68e9a8f108d02bc895b249","skill_md_path":"skills/blog-cannibalization/SKILL.md","default_branch":"main","skill_tree_url":"https://github.com/AgriciDaniel/claude-blog/tree/main/skills/blog-cannibalization"},"layout":"multi","source":"github","category":"claude-blog","frontmatter":{"name":"blog-cannibalization","license":"MIT","description":"Detect keyword cannibalization across blog posts by extracting primary keywords from titles and headings, clustering semantically similar targets, and flagging posts competing for the same search intent. Supports local-only mode (grep-based) and DataForSEO API mode (Page Intersection endpoint at ~$0.01/call). Outputs severity-scored report with merge or differentiate recommendations. Use when user says \"cannibalization\", \"keyword overlap\", \"competing pages\", \"duplicate keywords\", \"cannibalize\"."},"skills_sh_url":"https://skills.sh/AgriciDaniel/claude-blog/blog-cannibalization"},"updatedAt":"2026-05-18T18:53:29.541Z"}}