SuperAPI AI Directory
Partner with us中文AI newsFavorites
Skillsstrand-content-schema

strand-content-schema

Writing & research

View repository

Updated 10-15 17:29

About this skill

strand-content-schema

Every Strand post is an MDX file at content/posts/<slug>.mdx whose YAML frontmatter must satisfy PostFrontmatter (defined in packages/core/schema.ts). A post that does not validate cannot be committed or merged. Write frontmatter to pass on the first try, then self-check with strand validate <slug> or the MCP validate_post tool.

Required vs optional fields

Required: title, slug, description, publishedAt, author. Defaulted (safe to omit): status (draft), tags ([]), type (BlogPosting), noindex (false), faq ([]), sources ([]). Optional: updatedAt, canonicalUrl, featureImage, og, summary.

Rules that commonly trip validation

  • title ≤ 70 characters (SEO title length). Write for the SERP, not just the page.
  • description 50–160 characters (Bing SEO/GEO hard window is 25–160; authored posts keep the 50-char quality floor). Soft target 120–160 for SERP snippets. Make it a real summary with the primary keyword, not filler. @strand-cms/core also clamps via metaDescription() at emit time, but authored copy must still validate.
  • slug is kebab-case ^[a-z0-9]+(?:-[a-z0-9]+)*$. No spaces, capitals, underscores, or trailing hyphens. It must match the filename.
  • publishedAt / updatedAt are ISO 8601 datetimes.
  • author must reference an existing content/authors/<id>.mdx. Create the author file if the persona is new.
  • featureImage.alt is required when featureImage is present (accessibility + image SEO).
  • The body must not open with an # H1 heading — Strand themes render the frontmatter title as the page <h1>, so an in-body H1 ships a duplicate. Start the body with the answer-first lede paragraph; use ## for sections.

Get the GEO fields right (they are why this project exists)

  • summary — a ≤280-char TL;DR. It powers speakable schema and is the snippet AI search engines extract. Lead with the answer.
  • type — use NewsArticle for time-sensitive news, BlogPosting otherwise. This drives the JSON-LD type.
  • faq — 2–5 genuine question/answer pairs that match how people actually query. Emitted as FAQPage schema and frequently surfaced verbatim by AI search.
  • sources — every external factual claim should map to an entry here (see strand-fact-check-cite). Cited posts are trusted more by both Google and LLMs.

Example

---
title: "Bulgaria's 2026 EV Charging Grid: What Changed"
slug: bulgaria-2026-ev-charging-grid
description: "A plain-language look at Bulgaria's 2026 EV charging expansion: new fast-charger counts, grid funding, and what it means for drivers."
publishedAt: "2026-06-05T08:00:00Z"
author: maria-petrova
type: NewsArticle
tags: [bulgaria, ev, infrastructure]
summary: "Bulgaria added X fast chargers in 2026 under EU grid funding, concentrated on the A1/A2 corridors; rural coverage still lags."
faq:
  - q: "How many public EV chargers does Bulgaria have in 2026?"
    a: "Roughly N, up from M in 2025, with most growth on motorway corridors."
sources:
  - title: "Ministry of Energy 2026 infrastructure report"
    url: "https://example.gov.bg/report-2026"
    publisher: "Ministry of Energy"
---

After writing frontmatter, always run validation and fix any reported field before handing off to strand-publish.