Skip to content

sergekostenchuk/seo-llm-skill-cluster / skills/editorial-quality-gate

editorial-quality-gate

Verified

Review public site content for editorial quality, source support, prompt residue, duplicate blocks, translation drift, hidden filler, and human usefulness before SEO or LLM-friendly publication. Use this skill when the user asks to quality-gate news briefs, longform explainers, blog posts, topic pages, project pages, translated content, or generated article drafts without changing the live publishing pipeline.

references/editorial-quality-checklist.md

View on GitHub
# Editorial Quality Checklist

## Gate Decisions

| Decision | Meaning |
| --- | --- |
| `pass` | Publishable after normal copy edit. No severe factual, source, duplication, residue, or page-role issues. |
| `revise` | Useful content, but needs targeted fixes before publication. |
| `block` | Do not publish until major factual, source, duplication, translation, or residue issue is fixed. |

## Severity

| Severity | Examples |
| --- | --- |
| Critical | unsupported material claim, prompt residue, fabricated source, duplicate public article body, hidden SEO filler |
| Major | weak source framing, repeated context, unclear page role, literal translation causing meaning drift |
| Minor | awkward phrasing, missing context link, inconsistent terminology, weak summary |

## Source Support

Check:

- What happened?
- Who is involved?
- What changed?
- What source or artifact supports this?
- Is the claim stated as fact, inference, or opinion?
- Are dates and product/version names preserved?

Reject:

- "reportedly" without source;
- confident claims from a single weak source without framing;
- invented metrics;
- untraceable quotes;
- author opinions presented as external fact.

## Prompt Residue

Block or revise if public text contains:

- internal instructions such as "start with practical effect";
- planning comments such as "add reality check";
- meta commentary about how to write the article;
- leftover prompt labels;
- duplicated context from generation steps;
- "as an AI language model" or similar artifacts.

## Duplication And Content Model

For one-brief-plus-one-longform systems:

- brief is the short factual surface;
- longform is the explanatory surface;
- homepage/topic teasers summarize and route;
- do not create a third near-duplicate public story body;
- do not repeat the brief body verbatim across multiple index pages when a teaser can serve the page role.

## Translation Quality

Check:

- factual meaning preserved;
- target-language reader gets natural phrasing;
- product names and proper nouns are consistent;
- local idioms do not distort the technical meaning;
- dates and numeric claims are unchanged;
- source links and source titles remain traceable.

## Page-Type Checklist

### News Brief

- clear one-sentence event summary;
- source-backed claim;
- date visible;
- no unnecessary background padding;
- link to longform if available.

### Longform

- visible explanation of what changed;
- why it matters;
- source trail;
- uncertainties or limits;
- related links;
- no prompt residue.

### Blog

- author voice is clear;
- factual claims are supported;
- opinion is not disguised as news;
- useful takeaway for reader.

### Project

- purpose, status, audience, and artifacts are clear;
- repo/demo/docs links if public;
- limitations are not hidden;
- relation to site expertise is explicit.

### Topic

- direct definition;
- why this topic matters;
- representative stories/projects;
- no thin aggregation.
writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deploymentStars: 322kUpdated Claude CodeCodexVerifiedInstall
xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release.Stars: 118kUpdated Claude CodeCodexVerifiedInstall