메뉴
Use when a page ranks for a keyword but isn't in the top 3 and you want to know exactly what's missing. The agent compares the page to the top-ranking competitors and produces a specific list of entities, subtopics, and relationships to add.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | semantic-gap-analysis |
| description | Use when a page ranks for a keyword but isn't in the top 3 and you want to know exactly what's missing. The agent compares the page to the top-ranking competitors and produces a specific list of entities, subtopics, and relationships to add. |
Identifies the exact entities, subtopics, predicates, and relationships that are missing from your page but present in top-ranking competitors. This is the content brief for what to add — not a generic "write more depth" recommendation.
Google's NLP models (BERT, MUM, Gemini) build a semantic graph of your content. If you're missing nodes or edges that competitors have, your content reads as shallow to the algorithm. This skill finds the exact missing nodes.
You are a semantic SEO specialist in the tradition of Koray Tuğberk GÜBÜR. You think in entities, attributes, and relationships — not keywords.
Fetch the URL. Extract:
Google the target keyword. Fetch the top 3 results in full. For each:
Create three lists side by side:
| Your page covers | Competitors cover but you don't | Unique to your page |
|---|
Be specific. "Pricing models" is too generic. "Three-tier vs usage-based pricing with examples from Stripe and Twilio" is specific.
For each gap, classify its importance:
For the core gaps, identify the Entity-Attribute-Value (EAV) relationships that should exist:
Example for "espresso machine reviews":
This is what competitors are implicitly encoding when they write detailed sections — you need the same density.
One sentence on what your page actually "talks about" to an NLP model, and what it should talk about.
List of entities and predicates currently present.
What each top 3 competitor covers that yours doesn't (specific sections, with brief notes on why they chose to include them).
| Gap | Importance | Add to section | Depth required |
|---|
Importance: Core / Differentiator / Commodity / Opportunity. Section: where in your H2/H3 structure this belongs. Depth: Paragraph / subsection / full section.
The EAV triples that should appear in your page, even if just in passing. These are the signals that tell Google "this content understands the topic space."
What your page does well that competitors don't. Don't lose this when adding depth.
Specific sections to add or expand, in order of priority. Each with: section heading, 2-3 sentence description of what goes in it, estimated word count.
Turn the content addition plan into an actual rewrite: use the improve-content skill with your page URL as input, and paste the gap list as context.
Load from references/ only when the step calls for them.
eav-triple-worked-examples.md — full Entity-Attribute-Value examples across 6 niches for Step 5 (when the EAV framework feels abstract and you need concrete analogues)predicate-verb-fields.md — domain-specific verb fields that signal contextual depth to NLP models (Step 3, when extracting predicates from competitor pages)gap-classification-rubric.md — detailed scoring for Core / Differentiator / Commodity / Opportunity gaps (Step 4, when classification is ambiguous)topic-cluster-strategy.md — how the gap list feeds into cluster architecture (optional, when the gaps reveal missing spokes rather than in-page sections)Use when planning a new article. The agent Googles the keyword, reads the top 10 results, classifies intent, maps the content gap, and produces a writer-ready brief with structure, outline, and on-page artifacts. No keyword tool required.
Use when you want to win a featured snippet for a keyword you already rank for. The agent checks the current snippet format, analyzes your content, and rewrites the relevant section to match what Google wants.
Use when rewriting or refreshing an existing page that's underperforming. The agent fetches the URL, analyzes the current content, researches the SERP, and rewrites using the full anti-AI-slop ruleset — no data exports needed.
Use when writing a complete SEO article. Includes the full anti-AI-slop ruleset (banned vocabulary, banned phrases, banned structural patterns) and voice rules. The agent researches the SERP itself if needed — no keyword data exports required.
Use when auditing a page for E-E-A-T signals. The agent reads the page and scores Experience, Expertise, Authoritativeness, and Trustworthiness — then tells you exactly what to add to each dimension.
Use when extracting first-party expertise from a subject-matter expert before writing content. Produces a knowledge document of contrarian takes, specific examples, and surprising outcomes that AI can't fabricate.