Write a high-quality blog post from a marketing brief, PR, git diff, codebase feature, or freeform topic.
Detects the repo's existing blog voice (if there are prior posts), runs competitive research via web search, generates 3-5 headline options with explanations, and writes with proper SEO frontmatter — meta_description, primary_keyword, slug, secondary and long-tail keywords.
Three post types it can produce:
- Feature announcement — "We shipped X, here's why it matters" (~600-1000 words)
- Deep-dive technical — "How we built X" (~1500-2500 words)
- Problem-solution narrative — "You have problem X, here's how this solves it" (~1000-2000 words)
In resolution order:
- Path to a marketing brief (
.mdcontaining "Executive Summary" or "Key Messages") - GitHub PR URL or
#1234 - Git ref range —
v1.0...v2.0 - File or directory path
- Freeform text
/blog-post .tmp/marketing-brief.md
/blog-post #1234
/blog-post v1.4.0...v1.5.0
/blog-post src/features/auth
Or trigger by description:
"Write a blog post about the auth refactor in #1234."
Markdown file with SEO YAML frontmatter, saved to the auto-detected blog directory (content/blog/, _posts/, blog/posts/, etc.). Code examples, mermaid diagrams, and image placeholders with ready-to-use generation prompts are inserted inline.
Six phases: discovery (input + product context), competitive blog research (web search), configuration (audience, tone, post type, CTA, involvement level), outline approval, sectioned writing with self-review backfills, and headline selection. Three involvement levels — A (just write it), B (outline approval), C (section-by-section).
Pairs naturally with marketing-brief (write the brief first, feed it in), and with social-copy / newsletter downstream to amplify the post.