auto-content-blog
Full end-to-end SEO + GEO content creation pipeline for crypto/Web3 teams. Trigger this skill when the user wants to: write a blog post or article, research keywords, generate a content brief or outline, mine community comments for UGC insights, optimize content for SEO or AI search engines (GEO), score or audit an existing article, or check if writing sounds AI-generated. Also trigger automatically at the start of every new session to scan X for trending topics relevant to the user's project spec — propose hot keywords before the user asks. Accepts a project spec .md file as primary input. Also works with plain keyword or topic. Run all stages in sequence when starting from scratch. Enter mid-pipeline if user provides an existing draft or only asks for one stage.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/kaokray/cook-a-skill-content-blog-socialAuto Content blog
End-to-end pipeline: trending topic → keyword research → brief → UGC enrichment → human-style draft → visual & link enrichment → SEO + GEO optimization → scored QA report.
Pipeline
[0] X TREND MONITOR → auto-run on session start, propose hot keywords
[1] KEYWORD RESEARCH → scored, data-backed selection
[2] CONTENT BRIEF → outline + keyword map + UGC insertion points
[3] UGC ENRICHMENT → mine comments → ready-to-paste blog sections
[4] DRAFT → human-style, data-backed, zero AI patterns
[4.5] VISUAL & LINKS → data charts (matplotlib) + inline hyperlinks on every citation
[5] SEO + GEO → on-page checklist + AI-bot-friendly formatting
[6] QA REPORT → rubric score + publish verdict
Enter at any stage. Run all stages from scratch. Start at Stage 5 for existing drafts.
Quick commands:
run full pipeline→ Stage 0 → Stage 6start from stage [N]→ enter at any stagescore my draft→ Stage 5 → Stage 6enrich my draft→ Stage 4.5 only (charts + links on existing draft)
Reference files (read when detail is needed):
references/scoring-rubrics.md— all scoring tables (trend, keyword, UGC, SEO, GEO)references/ai-patterns-blacklist.md— full list of AI phrases to never writereferences/output-templates.md— exact output format for each stagereferences/example-run.md— real example: keyword "what is liquid staking"
Inputs
Required: project_spec (.md) + keyword_or_topic
Optional: ugc_urls (skip Stage 3 if absent), blog_draft, language (default: English), word_count (default: 1500–2000), tone (default: auto-detect from spec)
If project_spec is missing: stop and ask for it before starting.
Stage 0 — X Trend Monitor
Run automatically at the start of every new session, before the user says anything.
- Parse
project_spec.md→ extract: domain/niche, audience, product themes, competitor names - Run web searches (fill in values from spec):
site:x.com [domain keyword] -filter:replies[domain keyword] trending twitter 2025[domain keyword] discussion OR debate twitter[competitor name] twitter sentiment 2025crypto twitter trending today[domain keyword] CT crypto twitter discussion
- Score each topic → see
references/scoring-rubrics.mdStage 0 - Keep top 3. Output Trend Alert before any user message → see
references/output-templates.mdStage 0
Fallback if web search off: skip, say "Stage 0 unavailable. Provide a keyword to start Stage 1." Fallback if no trends found: list 3 evergreen topics from spec context.
Stage 1 — Keyword Research
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-kaokray-cook-a-skill-content-blog-social": {
"enabled": true,
"auto_update": true
}
}
}