Content Creation

Skopaq Content Creation

Create blog posts and changelog entries for the Skopaq public website. All content lives as MDX files in the Next.js dashboard, version-controlled in git, deployed via Vercel.

Content Locations

dashboard/content/blog/{YYYY-MM-DD}-{slug}.mdx       # Blog posts
dashboard/content/changelog/{YYYY-MM-DD}-{slug}.mdx   # Changelog entries

Workflow

1. Determine Content Type

2. Create MDX File

Every file requires complete frontmatter. Refer to references/frontmatter-template.md for the full template and field descriptions.

Critical rules:

• slug must match the filename (minus date prefix and .mdx)
• date must be ISO 8601 format (YYYY-MM-DD)
• category must be one of: engineering, benchmarks, changelog, case-study
• readingTime is estimated at ~250 words per minute
• coverImage path starts with /blog/ and the image must exist in dashboard/public/blog/
• author.avatar defaults to /icons/skopaq-logo.svg

3. Write Content Body

Voice and tone:

• Write in first-person plural ("we") for Skopaq Engineering posts
• Technical but accessible — assume the reader is a software engineer, not an AI specialist
• Lead with the outcome/result, then explain how
• Include real data — never use placeholder metrics

Structure for blog posts:

1. Opening hook (1-2 sentences with the key finding or announcement)
2. Context (why this matters)
3. Technical content (methodology, implementation, results)
4. Conclusion with forward-looking statement

Structure for changelog entries:

1. Version/date heading
2. Grouped by: Added, Changed, Fixed, Removed
3. Each item: one-line summary + optional detail paragraph

4. Use Custom MDX Components

Available components (defined in dashboard/components/blog/mdx-components.tsx):

Standard markdown elements (h1-h3, links, code blocks, tables, images) are styled automatically via the prose typography plugin.

5. Quality Checklist

Before committing any content:

• Frontmatter is complete (all required fields present)
• slug matches filename
• category is one of the 4 valid values
• excerpt is under 160 characters (SEO meta description length)
• readingTime is calculated (word count / 250, rounded up)
• No placeholder data — all metrics, dates, and claims are real
• Cover image exists at the specified path (or coverImage is omitted)
• Code blocks have language annotations (python, bash, etc.)
• Links use relative paths for internal pages (/blog/other-post)
• No "Argus" references — use "Skopaq" everywhere (per branding rules)
• Content renders correctly (cd dashboard && npm run build)

6. Verify Build

After creating content, always run:

cd dashboard && npm run build

This catches MDX parsing errors, missing imports, and broken frontmatter.

Additional Resources

Reference Files

• references/frontmatter-template.md — Complete frontmatter template with field descriptions and examples for both blog posts and changelog entries

Naming Conventions