suirindo
UserProduction-tested AIO/SEO toolkit for Next.js + Strapi + Japanese sites. Battle-tested on miyakodeit.com and netsujo.jp.
Categories
Indexed Skills (16)
ai-citation-tracker
Track brand and domain citations in ChatGPT/Perplexity/Gemini/AI Overviews over time. Issues structured queries, captures citation lists from each engine's response, persists snapshots, and reports week-over-week changes in citation share. Use when user says "AI citation", "AI流入", "ChatGPT引用", "Perplexity citation", "AI Overview mention", "brand visibility in AI", "GEO tracking", "share of AI voice", or "LLM SEO monitoring". Use proactively after content publication to measure GEO uplift.
bing-indexnow
Submit URLs to IndexNow (Bing/Yandex/Seznam/Naver) for instant indexation. Critical for AI search because 87% of ChatGPT Search citations and 100% of Microsoft Copilot citations come from the Bing index. Use when user says "IndexNow", "Bing indexation", "Copilot SEO", "ChatGPT Search SEO", "AI search indexation", "new article publish", "URL submit", "instant index" or after publishing/updating any blog article. Battle-tested on miyakodeit.com (51 sessions/90d via Bing).
ga4-custom-dimensions
Register the 12 standard custom dimensions in Google Analytics 4 via the Admin API. Idempotent, dry-run by default, YAML-driven catalog covering event-scoped, user-scoped, and item-scoped dimensions. Battle-tested on miyakodeit.com (GA4 property 531238165) and netsujo.jp (property 382871067). Use when user says "GA4 custom dimensions", "register dimensions", "customDimensions API", "event-scoped dimension", or "GA4 Admin API".
ga4-tracking-wiring
Wire GA4 event tracking (CTA clicks, outbound links, scroll depth, read completion) into a Next.js App Router project. Scaffolds a typed gtag wrapper, four event helpers, and four drop-in client components. Battle-tested on miyakodeit.com (PR
gsc-weekly-audit
Run a comprehensive Google Search Console audit covering Sitemap status, URL Inspection, Search Analytics, canonical mismatch detection, CTR diagnosis, and 0-impression page detection. Battle-tested on miyakodeit.com and netsujo.jp with weekly GitHub Actions cron. Use when user says "GSC audit", "Search Console weekly check", "indexing audit", "canonical mismatch", "CTR low pages", or wants automated site health monitoring.
h1-mobile-check
Detect Japanese H1 headings that break awkwardly on mobile (375px), where 1-2 characters spill onto a second line or particles/verbs get split mid-phrase. Renders the page at iPhone SE width, measures line boxes, and suggests optimal `<br />` insertion points using bunsetsu boundaries. Battle-tested on netsujo.jp and miyakodeit.com (3 prior incidents). Use when user says "H1", "mobile break", "見出し改行", "375px", "Japanese line break", "iPhone改行", "H1モバイル", or "見出しが崩れる".
indexation-recovery
Diagnose and recover "Discovered - not indexed" and "Crawled - not indexed" pages from Google Search Console. Pulls GSC coverage state, classifies the failure mode, proposes internal linking + content fixes, and triggers Indexing API submissions. Use when user says "indexed", "未インデックス", "クロール済み未インデックス", "検出未インデックス", "indexation", "coverage state", "Discovered", "Crawled not indexed", or "GSC coverage drop". Battle-tested on miyakodeit.com (recovered from 37% → target 60%+).
jp-ascii-space-fix
Detect and remove half-width spaces at ASCII↔CJK boundaries in Markdown, TSX/TS/JSX/JS, HTML, JSON-LD, and plain text. Context-aware — skips fenced code blocks, HTML attributes, and ASCII-only runs. Battle-tested across netsujo-web (178 files, 1500+ replacements in lib/blog-data.ts alone). Use when user says "half-width space", "CJK space", "Japanese typography", "和欧混植", "ASCII boundary", "英単語と日本語のスペース", or "remove space between English and Japanese".
jsonld-article
Generate and validate Article / BlogPosting / NewsArticle JSON-LD structured data for Next.js / Strapi / static sites. Auto-detects subtype, enforces Google rich results requirements (headline length, image aspect ratio, publisher logo size, ISO 8601 timezone), and produces drop-in React components. Battle-tested on miyakodeit.com blog (Article schema across all /blog/* posts, with ChatGPT and Perplexity citation hits). Use when user says "Article schema", "BlogPosting JSON-LD", "NewsArticle structured data", "article rich results", "blog post schema", "author markup", "publisher markup", or is adding JSON-LD to a blog/news/article page.
jsonld-breadcrumb
Generate and validate BreadcrumbList JSON-LD structured data for Next.js App Router, Strapi, and static sites. Auto-derives breadcrumb trails from URL pathnames via a label dictionary, outputs drop-in React components using usePathname(), and enforces Google rich results requirements (sequential position, absolute item URLs, omitted last-item URL). Battle-tested on miyakodeit.com `/study-group-kyoto` (PR
jsonld-event
Generate and validate Schema.org Event JSON-LD structured data for online, offline, and hybrid events. Supports OnlineEventAttendanceMode / OfflineEventAttendanceMode / MixedEventAttendanceMode, Google rich results requirements (timezone-aware startDate, 1920x1080 image), and outputs JSON / Next.js React component / Strapi v5 component. Battle-tested on miyakodeit.com (155+京都IT勉強会イベント) and used for Google Meet online meetups, Kyoto offline venues, and hybrid sessions. Use when user says "Event schema", "Event JSON-LD", "event structured data", "rich results event", "online event schema", "offline event schema", "hybrid event schema", "MixedEventAttendanceMode", "connpass schema", or has event listing pages that need rich results eligibility.
jsonld-faqpage
Generate and validate FAQPage JSON-LD structured data for Next.js / Strapi / static sites. Detects Q&A duplication across pages, validates against Schema.org spec, and produces drop-in React components. Battle-tested on miyakodeit.com (6 categories, 31 questions). Use when user says "FAQ schema", "FAQPage JSON-LD", "FAQ structured data", "rich results FAQ", or has multiple FAQ pages with overlap risk.
jsonld-organization
Generate and validate Organization JSON-LD structured data for Next.js / Strapi / static sites. Auto-expands sameAs from GitHub / X / LinkedIn / connpass / note handles, validates logo dimensions for Google rich results, and produces drop-in React components. Battle-tested on netsujo.jp as the citation source for ChatGPT search "Netsujo" queries. Use when user says "Organization schema", "Organization JSON-LD", "company schema", "about page schema", "brand schema", "sameAs", "logo schema", "rich results organization", or is wiring up `/company`, `/about`, or the site root for AI citation.
jsonld-speakable
Generate and validate WebPage + SpeakableSpecification JSON-LD structured data for Next.js / Strapi / static sites. Explicitly marks the passages that AI search engines and Google Assistant should read aloud or quote, raising AI citation confidence on ChatGPT, Perplexity, and AI Overviews. Battle-tested on miyakodeit.com (4 pages, `<p data-speakable>` markup synchronized with llms.txt). Use when user says "speakable schema", "SpeakableSpecification", "speakable JSON-LD", "AI Overviews speakable", "voice search schema", "AI citation schema", "data-speakable", "Google Assistant schema", or "llms.txt consistency check".
llms-txt-generator
Generate llms.txt and llms-full.txt for AI search optimization (GEO). Crafts AI-citable definition blocks, identifies key pages, and ensures consistency with Speakable JSON-LD across the site. Battle-tested on miyakodeit.com (cited by ChatGPT for "Kyoto IT 勉強会" query). Use when user says "llms.txt", "GEO", "AI search optimization", "ChatGPT citation", "AI Overviews", "Perplexity optimization", or wants to make their site AI-recommendable.
passage-citability-checker
Score the AI citation readiness of a page's first 200 tokens. AI Overview/ChatGPT/Perplexity cite 44.2% of passages from the first 30% of an article. This skill measures definition-density, statistic-density, answer-first structure, and entity clarity in the opening passage and proposes rewrites. Use when user says "AI引用", "GEO", "AI Overview optimization", "ChatGPT citation", "passage citability", "AIに引用される", "冒頭リード", or "intro rewrite".
Bio shown is the top-scored skill's repo description as a fallback — real GitHub bios land in a future update.