Claude Code Skill Treats Blog Posts Like Production Deploys
Because one-shot AI drafts routinely fail SEO and citation audits, this Claude Code skill suite runs every post through a 100-point delivery contract.

What it does
claude-blog is a Claude Code skill suite that orchestrates 30 sub-skills and 5 agents to research, write, and audit blog posts. Every draft ships as a folder containing Markdown, rendered HTML with JSON-LD schema, PDF, hero image, viewport screenshots, and a review scorecard. The system targets both traditional search rankings and AI citation platforms like ChatGPT and Perplexity.
The interesting bit
The v1.9.0 “5-gate Delivery Contract” acts like CI for content: a blog-reviewer agent scores drafts on a 100-point rubric and blocks anything below 90 from reaching you, with up to three auto-iteration loops before human escalation. It is a rare example of treating prose hygiene and SEO checklists with the same rigor as code review.
Key highlights
- 30 sub-skills organized into writing, strategy, research, and media clusters, dispatched through a central orchestrator.
- Dual optimization for Google rankings (E-E-A-T, schema markup, internal linking) and AI citations.
- Produces 8 output artifacts per post, including XSS-safe HTML, dark-mode-aware CSS, and responsive viewport screenshots.
- Version-coherence enforcement across 14 surfaces and prose-hygiene checks on every PR.
- Supports Next.js MDX, Astro, Hugo, Jekyll, WordPress, Ghost, 11ty, Gatsby, and static HTML output.
Caveats
- The README cites conflicting test counts (160 in prose, 187 in the badge) without clarifying which is authoritative.
- Documentation defaults to the private community mirror’s install flow; public users must mentally swap repo URLs and plugin slugs.
- The breadth of output artifacts (PDF, screenshots, hero images) suggests significant external dependencies, though the README does not detail their footprint.
Verdict
Worth exploring if you manage a publication pipeline inside Claude Code and need more rigor than a single-shot prompt. Skip it if you write only occasional READMEs or deeply technical documentation that does not target search or AI citation traffic.
Frequently asked
- What is AgriciDaniel/claude-blog?
- Because one-shot AI drafts routinely fail SEO and citation audits, this Claude Code skill suite runs every post through a 100-point delivery contract.
- Is claude-blog open source?
- Yes — AgriciDaniel/claude-blog is open source, released under the MIT license.
- What language is claude-blog written in?
- AgriciDaniel/claude-blog is primarily written in Python.
- How popular is claude-blog?
- AgriciDaniel/claude-blog has 1.4k stars on GitHub and is currently cooling off.
- Where can I find claude-blog?
- AgriciDaniel/claude-blog is on GitHub at https://github.com/AgriciDaniel/claude-blog.