AI constantly generates paradoxes \u2014 in work, learning, and training. The underlying question never changes: what value can we attribute to knowledge or experience when AI performs just as well, or better?<\/p>\n
For this post also, you can find all files, on my GitHub account. See https:\/\/github.com\/bflaven\/ia_usages\/tree\/main\/ia_seo_ia_semantic_breadcrumb_webmcp<\/a><\/b><\/p>\n On the subject of AI, I lean toward a clear-eyed pessimism. Each day confirms the erosion of the value of work. Productivity gains are so significant that the human effort historically invested in any given task is irrevocably disappearing. SEO is the most obvious example, and now GEO (Generative Engine Optimization) is following the same path. The practical conclusion of this post: hand it all over to AI and move on.<\/p>\n This brings to mind Gramsci’s well-known statement on the human condition:<\/p>\n \nIl mio stato d’animo sintetizza questi due sentimenti e li supera: sono pessimista con l’intelligenza, ma ottimista per la volont\u00e0.<\/em><\/p>\n “My state of mind synthesizes these two feelings and transcends them: I am a pessimist with my intellect, but an optimist by will.” \u2014 Antonio Gramsci\n<\/p><\/blockquote>\n Personally, I’m left with the pessimism. The will has partially deserted. There is little left to counter the abrasive effects of AI on work.<\/p>\n Take SEO \u2014 natural referencing. The shift from search<\/em> to response<\/em> is already underway. So what will remain of the audience\/advertising model after AI finishes dismantling it? This doesn’t only affect media companies; it applies to the entire digital economy and every platform that lives year after year on digital notoriety.<\/p>\n If audiences collapse and user interactions flow exclusively through chatbots or agents, nothing is left for anyone except AI operators \u2014 where competition is frankly unparalleled. The $65 billion raised by Anthropic and a valuation approaching a trillion dollars suggests the scale of the battles underway. But let’s get back to our modest topic: SEO.<\/p>\n Staying in the audience race \u2014 whether through SEO or GEO \u2014 is now non-negotiable. AI compliance may be the scythe that irrevocably decapitates the audience\/advertising model. Follow the logic to its conclusion:<\/p>\n Becoming AI-compliant, in this light, is a bit like feeding a viper in your bosom. The rot sets in. A bitter observation, perhaps exaggerated \u2014 fear inspired by AI is a poor advisor, and the worst-case scenario is never a certainty. But the question stands: how do we attract users to our platforms if AI answers on their behalf, drying up the entire click economy?<\/p>\n WebMCP is Google pulling the same old trick: dominant position, negative externality, and the bill sent to everyone else. The subtext, if you’ve invested heavily in SEO and organic search optimization, reads roughly like this: “Come save my business model \u2014 since you can’t save yours \u2014 by making your site compliant with a new WebMCP protocol, which is really just a more advanced JSON structure.”<\/em><\/p>\n A reminder on negative externalities: they occur when the cost of a transformation is imposed on the community without the initiating party bearing that cost. AI implementation is riddled with them:<\/p>\n We, as users, absorb all these negative externalities to stay competitive in the audience race \u2014 to the great benefit of an oligarchy of operators. Google’s strategy is transparent enough: shift from a results-engine model to an answer-engine powered by Gemini.<\/p>\n In short: from SERP (results) to chatbot (answers)<\/strong>. That’s the whole story.<\/p>\n The other thought that came to me while writing this post: declaring that AI makes everyone look like a fool and an incompetent is a dead end. The worst thing about this transformation is denying it. Some people confuse ends and means and see AI purely as a technological evolution, when it is in fact a source of unparalleled transformations.<\/p>\n The paradox: I’m inexorably absorbing all aspects of my digital product work \u2014 design, creation, execution \u2014 with the methodical help of AI. Does this diminish the value of my job or enhance my skills beyond what I could have imagined? I haven’t decided yet.<\/p>\n By establishing an abundance of cognitive capacity for everyone, doesn’t AI impoverish the overall thinking of each individual by eradicating scarcity \u2014 and therefore uniqueness? Consider artists. Look at speculation surrounding painters like Rothko, Basquiat, Warhol, Pollock. The moment death is announced, speculation runs rampant: the market knows no new works will be produced, so scarcity drives prices skyward. The Saatchi brothers built their fortune on exactly this mechanism \u2014 buying up entire outputs to control supply and dictate price.<\/p>\n This speculative mechanism is at the heart of the AI market: the destruction of human uniqueness, the appropriation of human creative value, and the resale of those capabilities \u2014 at exorbitant prices \u2014 to the very individuals who generated them in the first place.<\/p>\n Without verging too far into conspiracy theory (though I’m aware this is a fine line): the ongoing dumbing-down of individuals through entertainment, combined with the massive destruction of critical thinking via AI, seems to aim at something. As a Product Owner, I now create increasingly sophisticated proofs of concept \u2014 but I would be incapable of producing them without AI, and I find myself making fewer and fewer independent decisions. That’s my main concern.<\/p>\n On a positive note: with AI, mistakes become less costly. The anomaly is that errors and hesitations are now more valuable than results themselves \u2014 they are proof that we’re still thinking, at least a little. In the age of AI and Claude Code, the error represents rarity and therefore value, while rational accuracy is within reach of any fool. Remember Alex Karp’s quote from the previous post.<\/p>\n Enough. Let’s get back to something practical.<\/p>\n For those who manage a website and its data, the need is straightforward. Everything gathered below formed the matrix from which I built my prompts for Claude Code.<\/p>\n Through a Python pipeline, order WordPress post tags from taxonomies (e.g., The plugin must be able to simulate the breadcrumb, translation, enrichment, and slug before<\/em> validating any change to a tag from Since the breadcrumb is an ad hoc creation, changes can be stored in an additional MySQL table linked to the plugin. Preserve original elements at all times.<\/p>\n Proposed UX:<\/strong> A mapping view \u2014 original elements in the left column, modified elements in the right column \u2014 with all necessary fields to perform the optimized change. This allows generating the breadcrumb, translation, enrichment, and slug before pushing to production.<\/p>\n Renaming URLs using the label and slug requires a redirection table.<\/p>\n Everything is designed to generate a coherent breadcrumb. See the articulation of The plugin should allow, when creating tags in The real category taxonomy set up for this website. It was a game-changer for SEO and required the full pipeline to migrate.<\/p>\n Deleting the old category taxonomy required intervention in Common-sense advice \u2014 the kind you’d give to a talented student, or to an animal with superhuman strength. A few simple rules for managing AI, including Claude Code.<\/p>\n Install \u2014 Option A: Claude Code Plugin (recommended)<\/strong><\/p>\n From within Claude Code, add the marketplace first:<\/p>\n Then install the plugin:<\/p>\n This installs the guidelines as a Claude Code plugin, making the skill available across all your projects.<\/p>\n \nWebMCP is an open-source JavaScript library that allows any website to integrate with the Model Context Protocol. It provides a small blue widget in the bottom right corner of your page that allows users to connect to and interact with your webpage via LLM or agent.\n<\/p><\/blockquote>\n See: https:\/\/webmcp.dev\/<\/a><\/p>\n Keep an eye on it \u2014 given Google’s influence, this could become a de facto standard.<\/strong><\/p>\n Verdict: real but premature interest. Better to invest in AI constantly generates paradoxes \u2014 in work, learning, and training. The underlying question never changes: what value can we attribute to knowledge or experience when… <\/p>\nSEO Is Dead. Long Live GEO.<\/h2>\n
\n
WebMCP: Google Playing the Dominant Position Game (Again)<\/h2>\n
\n
The Philosophical Aside (Skip If You’re Impatient)<\/h2>\n
Tags Enrichment Pipeline for Thematic and Category Taxonomies<\/h2>\n
1. Pipeline Objective<\/h3>\n
taxonomy=post_tag<\/code>, taxonomy=category<\/code>) to build an extended breadcrumb \u2014 enriched for each tag with Wikidata information and named entity specification using Spacy. The final output is then handled through a WordPress plugin.<\/p>\n2. WordPress Plugin Objective<\/h3>\n
taxonomy=post_tag<\/code> or taxonomy=category<\/code> in production. This matters because it’s an automatically generated element that may require editorial approval.<\/p>\n3. Rules for the Enriched Breadcrumb<\/h3>\n
1. Primary Group \u2014 Category Tags (
taxonomy=category<\/code>)<\/h4>\n\n
Home \u203a AI<\/code><\/li>\n<\/ul>\n2. Secondary Group \u2014 Thematic Tags (
taxonomy=post_tag<\/code>)<\/h4>\n\n
Home \u203a AI \u203a Claude Code<\/code><\/li>\n3. Impact on URLs and Breadcrumb<\/h4>\n
\r\nURL: https:\/\/flaven.fr\/category\/ux-experience-utilisateur\/ (NOK)\r\nURL: https:\/\/flaven.fr\/category\/ux-user-experience\/ (OK)\r\nBreadcrumb: Home \u203a UX\r\n<\/pre>\n
taxonomy=category<\/code> and taxonomy=post_tag<\/code>.<\/p>\nWordPress Plugin Behavior<\/h4>\n
taxonomy=post_tag<\/code> and taxonomy=category<\/code>, the following manual operations:<\/p>\n\n
wikidata_id<\/code><\/li>\n\r\nHome \u203a Agile \u203a Claude Code (NOK)\r\nHome \u203a AI \u203a Claude Code (OK)\r\n<\/pre>\n
\n
\r\n# taxonomy=category\r\nURL: https:\/\/flaven.fr\/category\/agile\/\r\nBreadcrumb: Home \u203a Agile\r\nURL: https:\/\/flaven.fr\/category\/ux-experience-utilisateur\/\r\nBreadcrumb: Home \u203a UX\r\nURL: https:\/\/flaven.fr\/category\/tutoriaux\/\r\nBreadcrumb: Home \u203a Tutoriaux\r\netc...\r\n\r\n# taxonomy=post_tag\r\nURL: https:\/\/flaven.fr\/tag\/ai\/\r\nBreadcrumb: Home \u203a D\u00e9veloppement \u203a AI\r\nURL: https:\/\/flaven.fr\/tag\/wordpress\/\r\nBreadcrumb: Home \u203a WordPress\r\nURL: https:\/\/flaven.fr\/tag\/python\/\r\nBreadcrumb: Home \u203a Technologie \u203a python\r\nURL: https:\/\/flaven.fr\/tag\/php\/\r\nBreadcrumb: Home \u203a WordPress \u203a PHP\r\netc...\r\n<\/pre>\n
User Story<\/h3>\n
\r\nAs an editorial manager,\r\nI want a WordPress plugin to transform my site's taxonomy (tags and categories)\r\nfrom a flat structure (SEO-optimized) to a hierarchical and enriched structure (GEO-optimized).\r\n\r\nIn order to:\r\n- Manage tags and categories intelligently:\r\n - Detect named entities.\r\n - Enrich them via Wikidata.\r\n - Generate an optimized breadcrumb trail.\r\n - (Optional) Translate from a pivot language to the publication language.\r\n\r\n- Process existing tags in bulk (catch-up).\r\n- Manually correct tags after the catch-up.\r\n- Integrate new tags as they are created (anticipation).\r\n<\/pre>\n
Quick Must-Have Workflow Steps<\/h3>\n
\n
taxonomy=category<\/code> and taxonomy=post_tag<\/code>) of posts in French, with option to translate to English. Indicate ID and the number of content items attached to each.<\/li>\nExtra Info for the Plugin<\/h3>\n
\n
\r\n# Named entities from Spacy\r\nPERSON: People, including fictional.\r\nNORP: Nationalities or religious or political groups.\r\nFAC: Buildings, airports, highways, bridges, etc.\r\nORG: Companies, agencies, institutions, etc.\r\nGPE: Countries, cities, states.\r\nLOC: Non-GPE locations, mountain ranges, bodies of water.\r\nPRODUCT: Objects, vehicles, foods, etc. (Not services.)\r\nEVENT: Named hurricanes, battles, wars, sports events, etc.\r\nWORK_OF_ART: Titles of books, songs, etc.\r\nLAW: Named documents made into laws.\r\nLANGUAGE: Any named language.\r\nDATE: Absolute or relative dates or periods.\r\nTIME: Times smaller than a day.\r\nPERCENT: Percentage, including \"%\".\r\nMONEY: Monetary values, including unit.\r\nQUANTITY: Measurements, as of weight or distance.\r\nORDINAL: \"first\", \"second\", etc.\r\nCARDINAL: Numerals that do not fall under another type.\r\n<\/pre>\n<\/li>\n<\/ol>\n
Technical Environment for the Pipeline<\/h3>\n
\r\n- Conda: `tags_treatment`, Python 3.9.13\r\n- No Python 3.10+ syntax (no `match\/case`, no `X | Y` for types)\r\n- CSV separator: semicolon (`;`)\r\n- Source CSV encoding: `utf-8-sig` (BOM handling)\r\n- JSON encoding: `utf-8`\r\n- Logs: standard `logging` module, never `print()` in pipeline scripts\r\n- Dependencies: `pyyaml`, `requests`, `sentence-transformers`, `scikit-learn`,\r\n `spacy`, `streamlit`, `pandas`\r\n<\/pre>\n
Pipeline Working Commands<\/h3>\n
\r\n# step_1 \u2014 list tags\/categories\r\npython source\/pipeline\/001_step_1_list_tags_categories_wp.py \\\r\n --limit 30 \\\r\n --taxonomy post_tag \\\r\n --no-dry-run\r\n\r\npython source\/pipeline\/001_step_1_list_tags_categories_wp.py \\\r\n --limit 30 \\\r\n --taxonomy category \\\r\n --no-dry-run\r\n\r\npython source\/pipeline\/001_step_1_list_tags_categories_wp.py \\\r\n --limit 103 \\\r\n --taxonomy category \\\r\n --no-dry-run\r\n\r\n# step_2 \u2014 Spacy NER\r\npython source\/pipeline\/002_step_2_spacy_ner.py \\\r\n --limit 30 \\\r\n --taxonomy post_tag \\\r\n --no-dry-run\r\n\r\npython source\/pipeline\/002_step_2_spacy_ner.py \\\r\n --limit 2495 \\\r\n --taxonomy post_tag \\\r\n --no-dry-run\r\n\r\n# step_3 \u2014 Wikidata enrichment (use --limit carefully)\r\npython source\/pipeline\/003_step_3_wikidata_enrich.py \\\r\n --limit 30 \\\r\n --taxonomy post_tag \\\r\n --no-dry-run\r\n\r\npython source\/pipeline\/003_step_3_wikidata_enrich.py \\\r\n --limit 2495 \\\r\n --taxonomy post_tag \\\r\n --no-dry-run\r\n\r\n# step_4 \u2014 breadcrumb proposal\r\npython source\/pipeline\/004_step_4_breadcrumb_proposal.py \\\r\n --limit 30 \\\r\n --taxonomy post_tag \\\r\n --no-dry-run\r\n\r\npython source\/pipeline\/004_step_4_breadcrumb_proposal.py \\\r\n --limit 100 \\\r\n --taxonomy post_tag \\\r\n --no-dry-run\r\n\r\npython source\/pipeline\/004_step_4_breadcrumb_proposal.py \\\r\n --limit 2495 \\\r\n --taxonomy post_tag \\\r\n --no-dry-run\r\n<\/pre>\n
\r\n# Environment setup\r\ncd \/Users\/brunoflaven\/Documents\/01_work\/blog_articles\/_ia_seo_ia_semantic_breadcrumb\r\n\r\n# Conda\r\nconda create --name tags_treatment python=3.9.13\r\nconda info --envs\r\nsource activate tags_treatment\r\nconda deactivate\r\n\r\n# Docker (WordPress)\r\ncd \/Users\/brunoflaven\/Documents\/01_work\/blog_articles\/_ia_seo_ia_semantic_breadcrumb\/wp_docker\r\n\r\n# Start \/ Stop\r\n# docker compose down\r\n# docker compose up -d\r\n\r\n# Check\r\n# WordPress: http:\/\/localhost:8080\r\n# phpMyAdmin: http:\/\/localhost:8081\r\n<\/pre>\n
Production Category Taxonomy<\/h3>\n
\r\n\/* PRODUCTION *\/\r\n3438, \/\/ AI & Machine Learning\r\n3439, \/\/ APIs & Integration\r\n3437, \/\/ Business & Case Studies\r\n3440, \/\/ Cloud & Infrastructure\r\n3441, \/\/ Data & Analytics\r\n3436, \/\/ Digital Storytelling & Webdocs\r\n3454, \/\/ Journalism & Writing\r\n3453, \/\/ Miscellaneous \/ Other\r\n3442, \/\/ Mobile & Devices\r\n3443, \/\/ Multimedia & Video\r\n3444, \/\/ Programming & Databases\r\n3445, \/\/ SEO & Web Marketing\r\n3446, \/\/ Social Media & Community\r\n3447, \/\/ Technology & Trends\r\n3448, \/\/ Tools & Productivity\r\n3449, \/\/ Tutorials & How-to\r\n3450, \/\/ UX & Product Design\r\n3451, \/\/ Web Design & Front-end\r\n3435, \/\/ Web Development\r\n3452, \/\/ WordPress & CMS\r\n<\/pre>\n
.htaccess<\/code> since URLs were disappearing.<\/p>\n\r\n# BEGIN Breadcrumb Migration Redirect\r\n\r\nRedirect 301 \/category\/2-saas-gestion-dun-projet\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/5-gestion-des-textes\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/7-gestion-des-videos\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/3-saas-gestion-dune-sequence\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/6-gestion-de-laudio\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/8-gestion-des-boutons-et-des-actions\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/8-saas-gestion-des-effets\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/9-saas-gestion-des-evenements\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/4-gestion-des-images\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/4-saas-gestion-des-images\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/91-saas-gestion-des-medias-13\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/6-saas-gestion-des-sons\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/5-saas-gestion-des-videos\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/92-saas-gestion-du-style\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/3-configuration-dune-sequence\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/nos-derniers-wdoc\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/1-saas-mon-compte\/ \/category\/3wdoc-tutorials\/\r\nRedirect 301 \/category\/1-mon-compte-sur-3wdoc\/ \/category\/3wdoc-tutorials\/\r\n\r\n# END Breadcrumb Migration Redirect\r\n<\/pre>\n
Karpathy-Inspired Claude Code Guidelines<\/h2>\n
\n
\r\n\/plugin marketplace add forrestchang\/andrej-karpathy-skills\r\n<\/pre>\n
\r\n\/plugin install andrej-karpathy-skills@karpathy-skills\r\n<\/pre>\n
\n
More on WebMCP<\/h2>\n
What Is WebMCP in a Nutshell<\/h3>\n
\n
Claude’s Assessment (Honest Summary)<\/h3>\n
\n
llms.txt<\/code> + clean semantic markup. WebMCP plugin = over-engineering for most current use cases.<\/li>\nllms.txt<\/code> + per-agent robots.txt<\/code> directives + a clear AI licensing policy first.<\/strong><\/p>\nWebMCP Resources<\/h3>\n
\n