You can talk to us before requirements are fixed.
Consultation and quotes are free. We reply by the next business day.
AI Content Quality Engine
Kirico
AI can write articles.
Kirico gets them ready to publish.
Generate with RAG from your existing articles, inspect with Quality Gate, prepare SEO/social/translation assets, and draft into microCMS. Editors focus on the publish decision. Legal-risk phrasing checks are meant to run before go-live.
Official microCMS PartnerWhat does KIRICO stand for?
Knowledge
RAG-powered reference search
Inspection
Legal compliance & quality gate
Review
LLM-powered contextual review
Intelligence
Coverage, gaps & priorities
Content
Everything we work with
Like the Japanese art of Kiriko glass cutting — carefully polishing each surface to perfection. That's what Kirico does for your content.
Common Challenges We Solve
- Manually setting SEO and posting to social media for every article
- Spending too much time on content quality checks
- Worried about missing legal compliance risks
- Want multilingual support but translation costs are a burden
- Don't know what's missing compared to competitors
- Can't decide which existing articles to rewrite first
Benefits
Reduce Operations by 80%
Automate manual work around publishing — SEO setup, OGP creation, social posts, and translation
Catch quality risks early
AI automatically checks for legal compliance risks before publishing, before go-live
Evidence over guesswork
Coverage analysis, competitive gaps, and rewrite priorities replace guesswork with evidence
Features
AI Content Generation
Generate articles from themes and keywords. SEO metadata, social media posts, and OGP images are produced simultaneously. RAG-powered reference search ensures content builds on your existing assets.
- AI Article Generation
- Automated SEO Title & Description
- Social Posts for 4 Platforms
- Dynamic OGP Image Generation (1200×630)
Quality Inspection & Governance
Two-stage quality assurance: instant regex checks plus LLM-powered contextual review. Automatically detects legal risks. Supports custom rules per project.
- Legal Risk Detection (Advertising & Pharmaceutical)
- Terminology & Tone Consistency
- LLM-powered Logic & Fact Checking
- One-click Auto-fix
Content Automation Pipeline
Simply publish in microCMS and the rest happens automatically. Webhook integration eliminates manual work for quality checks, SEO setup, social distribution, and translation.
- microCMS Webhook Integration
- Auto Quality Check on Publish
- Multilingual Translation (EN/ZH/KO+)
- Auto-draft to microCMS
Content Intelligence
Visualize content coverage, analyze competitive gaps, and prioritize rewrites with data-driven scoring.
- Category-based Coverage Analysis
- Competitive Gap Detection
- Rewrite Priority Scoring
- Content Map Visualization
Automation Pipeline
Publish
As usual in microCMS
Webhook
Auto-notify Kirico
Quality Gate
Legal & rule checks
AI Pipeline
SEO, OGP, SNS, i18n
Delivered
Auto-distributed
Before & After
| Item | Before | Kirico |
|---|---|---|
| Writing | Manual or generic AI paste | RAG from existing articles |
| Review | Ad-hoc human checks | Quality Gate publish control |
| SEO / Social / i18n | Manual after drafting | Prepared with generation |
| CMS entry | Copy-paste | Draft into microCMS |
| Rewrite priority | Guesswork | Data-driven scoring |
Packages
Not a flat SaaS fee — we propose PoC, production rollout, and training step by step. Pricing is quoted per scope.
Kirico PoC
RecommendedAbout 2–4 weeks
Custom quote
Validate generation through draft publish on one site / endpoint.
- microCMS connection
- Crawl existing articles
- RAG generation demo
- Initial quality rules
- Draft publish check
- Improvement report
Production pack
Go liveOngoing ops design
Custom quote
Webhooks, rule design, manuals, and training for production.
- Per-project setup
- Webhook integration
- Quality rule design
- Ops manual
- Team training
- Go-live support
Workshop
Learn as a teamHalf day to 1 day
Custom quote
Hands-on workshop on microCMS × AI design and governance.
- AI-era CMS design
- Schema / Webhook
- Quality Gate hands-on
- AIO / LLMO inventory
- Own-content exercise
Process
Discovery
Assess your microCMS usage and content workflows to identify automation opportunities with Kirico.
Design & PoC
Design quality rules and build AI pipelines. Validate effectiveness with a small-scale proof of concept.
Deployment
Connect Kirico to production. Set up microCMS webhooks and apply custom quality rules.
Optimize
Continuously tune quality rules, improve inspection accuracy, and add new checks as needs evolve.
Technologies
CMS
AI / LLM
Quality
Automation
Related Services
Pricing guide
Prices are pre-tax guides and vary with scope, integrations, and post-launch operations. Formal quotes are free.
Related work
Projects close to this service. Those with challenge, scope, and stack described come first.
Case study株式会社Nature Innovation Group
傘レンタルアプリ「アイカサ」
Built the umbrella-sharing service from scratch in its early phase
Scope|デザイン / コーディング / システム開発
株式会社ペイミー
Payme（ペイミー） LPサイト
Scope|デザイン / コーディング / マーケティング
Case study学校法人松本学園（広島桜が丘高等学校・広島県瀬戸内高等学校）
学食モバイルオーダー
Students pre-order and pay on their phones, easing cafeteria queues and cash handling
Scope|デザイン / コーディング / システム開発
AI Search (AIO / LLMO)
What are LLMO and AIO, and how do they differ from SEO?
LLMO (AIO) is the work of making sure generative AI and AI search — ChatGPT, Gemini, Claude — understand your company correctly and describe it accurately. SEO is about ranking; LLMO is about whether an AI answer includes you as an option. Being recommended as a candidate tends to matter more for results than merely being cited as a source.
Can you check how AI tools currently describe our company?
Yes. We run the questions your buyers would actually ask ("which development firm is strong in X?", "how is company A rated?") and record whether you appear and how you are described. Where the description does not match reality, we trace it back to the wording on your own site and fix that. We can do a quick version of this in the first meeting.
How do you measure the impact of AI search?
No single metric captures it, so we combine several: the share of inquiries that report "found via an AI answer" in the how-did-you-hear-about-us question, branded search volume, the recommendation rate for a fixed set of prompts, and how closely AI descriptions match reality. Our own contact form asks that attribution question.
Where should we start with AI search?
Start by answering the questions your customers are likely to have, on your own site. Writing out 20–30 likely questions and answering them is already a solid first step. Next, rewrite vague claims ("we have deep expertise") into concrete facts: number of clients, years, industries, scope. AI can only quote what is stated specifically.
Does our CMS or content structure affect AI search?
It does. Structured, field-level content is easier for AI to reference, and you can reuse the same content as a retrieval source (RAG) internally. As an official microCMS partner we help design that structure and automate the operations around it. Sites with stale content lose both search evaluation and accuracy in how AI describes them.
Do we need llms.txt and structured data (JSON-LD)?
They are worth having: they make your site machine-readable for AI and crawlers. Our own site publishes llms.txt and per-page structured data. But they only expose what is written — if the content itself is vague, the markup will not help. We do both together.
You can talk to us before requirements are finished.
We can start by sorting what to do first. We run businesses ourselves, so the conversation stays about labor and ops. Consultation and estimates are free.
Online meetings / reply by next business day / free consult & estimate