ghost-content-engine

Build and operate a self-hosted Ghost content engine that turns YouTube videos, meeting notes, community wins, quote banks, Neo4j graph context, and Reddit exact-question FAQs into SEO/AEO-friendly Gh

SKILL.md

GitHub

Ghost Content Engine

Use this skill when the user wants to create, enrich, publish, or maintain blog posts for a self-hosted Ghost site using the Ghost Content Engine Kit.

Workflow

  1. Read .env.example and confirm .env has GHOST_URL, GHOST_ADMIN_API_KEY, SITE_NAME, and SITE_AUTHOR_NAME.
  2. For Ghost publishing, use scripts/ghost_client.py.
  3. For YouTube videos, use scripts/youtube_blog_pipeline.py.
  4. For Reddit FAQ market language, use scripts/reddit_question_mining.py.
  5. For internal graph enrichment, use scripts/neo4j_question_mining.py.
  6. For combined FAQ enrichment, use scripts/faq_enrichment_pipeline.py.
  7. For community wins, case studies, masterclasses, newsletters, or quote banks, use scripts/community_blog_pipeline.py.
  8. For LinkedIn repurposing from a Ghost post, use scripts/ghost_to_linkedin.py or scripts/linkedin_post_pipeline.py.

Guardrails

  • Never hardcode API keys.
  • Use --dry-run before live Apify runs.
  • Keep Reddit question wording exact, but answer in the site owner's voice.
  • Keep private community names and links redacted unless permission is explicit.
  • Publish through Ghost HTML cards with --html-card so custom classes survive.
  • Keep important content visible in HTML; do not hide essential content behind JavaScript.
  • Use clean evergreen slugs.
  • LinkedIn publishing must use official OAuth/API credentials and must not use cookies or browser automation.
  • Use --review-only or --dry-run before --publish-now unless the user explicitly asked for live posting.

Common Commands

python scripts/ghost_client.py list-posts --limit 5
python scripts/youtube_blog_pipeline.py from-url --url "https://www.youtube.com/watch?v=VIDEO_ID" --faq-dry-run
python scripts/reddit_question_mining.py research --topic "AI agency first client" --slug ai-agency-first-client --dry-run
python scripts/neo4j_question_mining.py research --topic "AI agency first client" --slug ai-agency-first-client --max-nodes 500
python scripts/community_blog_pipeline.py from-file --source-file examples/community-win-source.md --title "How A Simple Website Offer Became A First-Client Win" --slug simple-website-offer-first-client-win --content-type win --redact "Member Name"
python scripts/ghost_to_linkedin.py --slug ai-agency-first-client --review-only
python scripts/ghost_to_linkedin.py --slug ai-agency-first-client --publish-now

References

Read the repo docs only when needed:

  • docs/setup.md
  • docs/youtube-repurposing.md
  • docs/reddit-faq-enrichment.md
  • docs/neo4j-content-memory.md
  • docs/community-content-pipeline.md
  • docs/publishing-checklist.md
  • linkedin-setup/README.md
  • linkedin-setup/cli-workflow.md