From content-engine
Validate blog articles against a structured quality checklist. Use when the user asks to "review an article", "check quality", "validate content", "run quality checklist", "score this article", or needs to verify an article meets publishing standards before going live.
npx claudepluginhub jeffrey94/jeffrey-skills --plugin content-engineThis skill uses the workspace's default tool permissions.
Dual-tier validation framework for blog articles. Choose the tier based on content type, then score against the checklist.
Audits and scores blog posts on 100-point system across content quality, SEO, E-E-A-T, technical elements, AI citation readiness. Detects AI content via burstiness/phrasing, supports batch analysis, prioritized fixes, exports (markdown/JSON/table).
Evaluates marketing content quality across six dimensions: content quality, brand voice, hallucination risk, claim verification, structure, readability. Scores drafts, flags issues with fixes before publication.
Audits content for quality, E-E-A-T signals, readability, keyword usage, structure, trust indicators, and SEO best practices with scores and recommendations.
Share bugs, ideas, or general feedback.
Dual-tier validation framework for blog articles. Choose the tier based on content type, then score against the checklist.
Tier 1 — Government Content (15-point checklist, pass: 13/15 = 86.7%) Use for: Policy, compliance, government schemes, government financing programs.
Tier 2 — Market Content (12-point checklist, pass: 10/12 = 83.3%) Use for: Market trends, success stories, operational guides, platform comparisons.
Load quality thresholds from config/project.json → content.quality_thresholds.
Check the first 2 paragraphs for a local currency value, percentage, or data point from a credible source.
Search for local currency values, percentages, and numerical data with sources. Prefer 5-7 distinct data points. Each must have a hybrid citation. Data should be current (within the past year).
At least 1 government program named with: official name, what it offers, deadlines, portal URLs. Load known programs from config/project.json → regulatory.
Article must target specific industries (not generic "all SMEs"). Check that named industries appear at least 3 times. Load target industries from config/project.json → audience.industries.
Must include:
Verify:
config/project.json → market.currency), never USD or genericAll bracketed citation letters [a], [b], [c] listed at end of article with source name and URL.
Immediately after hook. 3-5 bullet points with bold labels. Specific data (currency/percentage/deadline). Active voice, scannable.
5-7 questions near end, before CTA. Must include: 1× "What…", 1× "How…", 1× "When…/deadline", 1× "How much…/amount". Direct answers 2-4 sentences each.
At least 1 table if data warrants it. Use for: comparisons, multi-attribute data, step-by-step processes.
Minimize passive constructions. Check for: "can be", "was", "were", "is being" + past participle. Address reader directly with "you" and "your".
Conversational but professional. Clear, simple explanations. Confident expertise. No corporate jargon, no aggressive sales language, no generic platitudes.
Same as Tier 1 with these differences:
Any of these means automatic failure regardless of score:
Tier 1: 15/15 Excellent → 14/15 Great → 13/15 Good (pass) → ≤12/15 Needs revision Tier 2: 12/12 Excellent → 11/12 Great → 10/12 Good (pass) → ≤9/12 Needs revision
Present results as:
ARTICLE REVIEW
==============
Tier: [1 or 2]
Score: [X]/[15 or 12] ([percentage]%)
Status: [APPROVED / NEEDS REVISION]
CHECKS:
1. Hook with specific data [PASS/FAIL] — [brief note]
2. Statistics (3+) [PASS/FAIL] — [count found]
...
FEEDBACK:
- [Specific issue and how to fix it]
- [Another issue]
RECOMMENDATION:
[Approve for publishing / Revise with specific guidance]
For detailed check specifications, see references/checklist-details.md.