sergekostenchuk/seo-llm-skill-cluster / skills/editorial-quality-gate
editorial-quality-gate
VerifiedReview 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.
Editorial Quality Gate
Use this skill after the site content model and publication surface are known.
Read references/editorial-quality-checklist.md before producing a full content quality report.
Owns
- editorial quality checklist;
- source support and unsupported-claim review;
- prompt residue detection;
- repeated block and context leakage review;
- no SEO-filler checks;
- page-type quality gates for news, longform, blog, project, and topic pages;
- handoff backlog for content strategy, localization, and decay monitoring.
Does Not Own
- live publishing pipelines;
- model provider credentials;
- translation implementation;
- content strategy beyond quality-gate findings;
- content decay automation;
- technical SEO/schema implementation;
- visual layout or UX design.
Workflow
- Identify page type:
news_brief,longform_article,blog_post,project_page,topic_page, ortranslation. - Identify the intended source material and visible claims.
- Check whether each material claim is supported by a source, artifact, or clearly marked author opinion.
- Scan for prompt residue, editorial instructions, repeated context blocks, boilerplate padding, and low-value SEO filler.
- For translations, compare meaning and audience fit rather than demanding literal phrasing.
- Check that the content serves its declared page role and does not create a duplicate public story body.
- Return a gate decision:
pass,revise, orblock. - Record follow-up items for content strategy, localization, or decay monitoring when they exceed this skill's scope.
Non-Negotiables
- Do not approve unsupported factual claims.
- Do not approve hidden SEO-only content.
- Do not approve prompt residue or internal editorial instructions in public text.
- Do not approve repeated paragraphs unless repetition is explicitly justified by page structure.
- Do not create a third public story body when the declared model is one short brief plus one longform article.
- Do not claim translation quality without checking meaning, terminology, tone, and locale fit.
- Do not rewrite the whole article unless the user explicitly asks for a rewrite.
- Do not touch live publishing systems.
Safety And Privacy Boundaries
- Do not include secrets, API keys, private analytics rows, raw IP logs, account tokens, private drafts, or admin-only URLs in public content reports.
- Do not publish or edit live content unless the user separately asks for that implementation task.
- Do not reveal private source material in public-facing recommendations.
- Do not approve claims about medical, legal, financial, ranking, crawler, or assistant-citation outcomes without appropriate evidence and scope labels.
- Do not turn editorial QA into hidden crawler-only optimization.
Page-Type Gates
News Brief
Must be concise, sourced, dated, and clear about what happened. It should link to the paired longform when available. It should not become a longform explanation.
Longform Article
Must explain context, why it matters, source trail, implications, and uncertainty. It should remove prompt residue from generation and avoid repeating the short brief verbatim beyond a short summary.
Blog Post
May contain author opinion, but factual claims and references still need support or clear framing. It should not imitate news authority if it is a personal essay.
Project Page
Must describe purpose, status, repository or artifact links if public, evidence of usefulness, and relation to the site or workflow. It should not overclaim maturity.
Topic Page
Must define the topic, explain why it matters, link to representative stories/projects, and avoid thin aggregation.
Translation
Must be localized for the target reader, preserve factual meaning, avoid literal awkwardness, and keep names, dates, product terms, and source references consistent.
Output Shape
Return:
- Gate decision:
pass,revise, orblock. - Page type and intended role.
- Findings grouped by severity.
- Source-support notes.
- Prompt-residue and duplication notes.
- Translation/localization notes if relevant.
- Minimal edits or revision instructions.
- Deferred backlog items for content strategy, localization, or decay monitoring.
Use assets/quality-gate-report.template.md when a persistent report is needed.
Validation
Validate skill edits with:
python3 <codex-skills-dir>/senior-skill-architect/scripts/lint_production_skill.py ./skills/editorial-quality-gate
python3 ./plans/seo-llm-skill-cluster/scripts/lint_skill_cluster.py .
Forward tests live in evals.json.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deployment | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| 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: 198k | Updated | Claude CodeCodex | Verified | Install |
| 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: 119k | Updated | Claude CodeCodex | Verified | Install |
| 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: 118k | Updated | Claude CodeCodex | Verified | Install |