# Syted: complete reference

Syted is a subscription service that researches, writes and publishes SEO
articles for SaaS companies. The articles are served through an authenticated
API and rendered by the customer's own application, so the blog lives at
yourdomain.com/blog rather than on a hosted subdomain.

Website: https://syted.ai
Price: $99 per month, 30 articles per month
Package: @syted/next on npm

---

## 1. The problem it solves

Search results for the same phrase can serve entirely different audiences. A
keyword with 1,600 monthly searches, a difficulty of 5 out of 100 and an $8
cost per click looks like an obvious target on any keyword tool. Opening the
results can reveal that every page ranking for it is an agency reporting
dashboard, a white-label PDF builder or a Looker Studio template, none of which
sells what the customer sells.

Volume and difficulty describe a query. They say nothing about who is asking or
what they expect to find. That information exists only on the results page.

Syted reads that page before writing. When what ranks does not match what the
customer sells, the keyword is skipped and the reason is recorded. A skipped
article is the intended behaviour.

## 2. How an article is produced

1. Keyword research runs every two weeks and fills a queue of candidate topics.
   Running it per article would multiply the data cost with no gain, since the
   underlying metrics move slowly.
2. Every morning, one topic leaves the queue. The live search results for it are
   fetched fresh, including the AI Overview when there is one.
3. The intent check compares what ranks against the customer's own description
   of their product and audience. A mismatch stops the run before any tokens are
   spent on writing.
4. A brief is built from what actually ranks: the questions people ask, the
   sections competitors cover, the angle nobody has taken.
5. The article is written against that brief and against the list of facts the
   customer allows the writer to state.
6. A separate review pass reads the draft against the brief, the fact list and
   the quality rules. It scores the draft and reports what is wrong.
7. Articles that pass are published. Articles that fail are stored as drafts and
   surfaced in the dashboard, never pushed to the customer's site.

## 3. What the customer configures

- Product description and what it does
- Who buys it, in their own words
- Facts the writer may state: prices, plan names, limits, published numbers
- Phrases that must never appear
- Tone and reading level
- Internal pages worth linking to
- Competitors, so their names are handled deliberately rather than by accident

Nothing about the product is inferred. A claim outside the fact list is not
written, which is what stops an article inventing a price the customer does not
charge.

## 4. LLM visibility

Syted measures how often the customer's domain appears in ChatGPT answers,
weekly, and which domains are cited instead of them on the same questions. That
second list is usually more useful than the first: it is the real competitive
set for the answer, and it is rarely the set the customer assumed.

No guarantee of a recommendation is offered, because none is possible. The
number is reported and tracked over time.

## 5. Integration

### Install

    npm i @syted/next

Set the API key in the server environment. It must never reach the browser.

    SYTED_API_KEY=ab_live_...

### Turnkey mode

One file renders the whole blog.

    // app/blog/[[...slug]]/page.tsx
    import { createBlogPage } from '@syted/next'

    const blog = createBlogPage({
      title: 'Blog',
      siteUrl: 'https://acme.com',
      headerOffset: 64,
      cta: {
        eyebrow: 'Acme',
        title: 'Ship your changelog in minutes',
        description: 'Free while you are under 100 users.',
        buttonLabel: 'Start free',
        href: '/signup',
      },
    })

    export default blog.Page
    export const generateStaticParams = blog.generateStaticParams
    export const generateMetadata = blog.generateMetadata
    export const revalidate = 3600

It renders the index, the article pages, canonical URLs, Open Graph and Twitter
metadata, BlogPosting and BreadcrumbList JSON-LD, a reading progress bar pinned
to the top of the window, a sticky table of contents built from the article's
own headings, a CTA card in the right rail and a CTA banner at the end of the
article.

Options: title, description, basePath (default /blog), siteUrl, cta, footerCta,
toc (default true), tocLabel, headerOffset in pixels, theme ('light' or 'auto'),
pageSize, and classes for root, index, indexTitle, post, postTitle and prose.

No CSS framework is required. Styles are injected by the component and driven by
CSS variables on .ab-root: --ab-fg, --ab-muted, --ab-faint, --ab-border,
--ab-hairline, --ab-surface, --ab-accent, --ab-radius, --ab-max, --ab-font.
Redefining one in the host stylesheet overrides the package.

### Data-only mode

For teams who render their own design.

    import { createClient } from '@syted/next/client'

    const blog = createClient()
    const { articles, total } = await blog.list({ limit: 20, tag: 'seo' })
    const article = await blog.get('some-slug')   // article.body is Markdown
    const { tags } = await blog.tags()
    const entries = await blog.sitemap()

Building blocks are exported separately:

    import { renderArticle, ReadingProgress, TableOfContents } from '@syted/next'

    const { html, headings } = renderArticle(article.body)
    // headings: [{ depth: 2 | 3, text, id }]

The ids in the HTML match the ids in the headings array, so a custom table of
contents resolves without extra work.

### Instant updates

    // app/api/syted/revalidate/route.ts
    export { POST } from '@syted/next'

    SYTED_WEBHOOK_SECRET=...

Syted calls this route on publish so the new article appears immediately
instead of waiting for the cache to expire.

### Sitemap

    // app/sitemap.ts
    import { createClient } from '@syted/next/client'

    export default async function sitemap() {
      const entries = await createClient().sitemap()
      return entries.map((e) => ({
        url: e.loc,
        lastModified: e.lastmod,
        changeFrequency: e.changefreq,
        priority: e.priority,
      }))
    }

## 6. Public API

Base URL https://syted.ai/api/v1, bearer token authentication with the site
API key.

    GET /articles?limit=&offset=&tag=&locale=
        -> { articles, total, limit, offset }
    GET /articles/:slug
        -> { article }  with Markdown body and ready-to-inject jsonLd
    GET /tags
        -> { tags: [{ name, count }] }
    GET /sitemap
        -> { entries: [{ loc, lastmod, changefreq, priority }] }

401 means the key is invalid or the subscription is no longer active.

## 7. Pricing and terms

- $99 per month, a single plan
- 30 articles per month
- Keyword research and live SERP verification included; no separate SEO tool
  subscription is needed
- Cancel in one click from the dashboard, with no retention flow and no call
- After cancellation the articles stop being served and disappear from the site.
  They are not exported. This is stated on the pricing page rather than left to
  be discovered later.

## 8. What Syted is not

It is not an SEO reporting or rank tracking tool. It does not audit sites, build
backlinks, manage ad campaigns or write on demand from a prompt. It researches,
writes and publishes articles on a schedule.