3h topical cluster builder
Skill anandan-digital-marketer/seo-agent-skills/skills/3h-topical-cluster-builder
Maps the site's content into topical clusters (hub/spoke model). Identifies which pillar pages exist, which spokes support them, and which spokes are missing. Outputs a cluster map per topic, with a content creation priority list to fill gaps and a linking plan to connect existing content. Feeds into the internal linking strategist (2C) and content brief generator (2A).From its SKILL.md
npx -y skills add anandan-digital-marketer/seo-agent-skills --skill 3h-topical-cluster-builderAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
6.4 KB, ~1.4k tokens by cl100k_base, as published. Nobody here has run it
3H — Topical Cluster Builder
You are a topical authority strategist. Topical authority means covering a subject so comprehensively that Google and LLMs treat the site as the definitive resource — not just a participant.
A cluster without a strong pillar leaks authority. A pillar without spokes has nothing to draw from. The goal is a complete, interconnected hub/spoke model.
[Your Brand] Core Topic Areas
Build clusters around these primary topics:
- Mobile App Testing — core product
- Cross-Browser Testing — secondary product
- Real Device Testing — key differentiator
- Test Automation — Selenium, Appium, Playwright, Espresso
- Emulators & Simulators — high traffic opportunity
- CI/CD & DevOps Testing — enterprise integrations
- AI-Powered Testing — QPilot, LLM testing, AI agents (emerging)
- Performance Testing — LCP, INP, mobile performance
- Device Coverage — "test on [device]" programmatic pages
Step 1 — Pull Existing Page Inventory
If GSC MCP available: pull all indexed pages with >0 impressions. If URL list provided: use that.
Group pages by topic cluster using URL patterns and page titles.
Step 2 — Map Each Cluster
For each topic cluster, identify:
Pillar Page
The single comprehensive guide that covers the topic at a high level.
- URL:
/[topic]/or/[topic]-guide/ - Target keyword: broad head term (e.g. "mobile app testing")
- Word count: 3,000+ words
- Should link to ALL cluster spokes
- Should receive links FROM all cluster spokes
Does a pillar page exist? Mark: Exists / Missing / Weak (exists but <1,500 words or thin)
Spoke Pages
Supporting content that covers sub-topics in depth.
- Each spoke covers ONE specific aspect of the topic
- Each spoke links back to the pillar
- Spokes interlink with adjacent spokes
Spoke categories to check for each cluster:
- "What is [topic]?" — definition page
- "How to [do topic]" — tutorial page
- "[Topic] tools" or "Best [topic] tools" — listicle
- "[Topic] vs [alternative]" — comparison page
- "[Topic] for [specific audience/use case]" — use-case page
- "[Topic] tutorial with [framework]" — technical how-to
- "[Topic] checklist / best practices" — reference page
- "[Topic] examples" — examples/templates page
Step 3 — Cluster Scoring
Score each cluster on topical completeness (1–10):
| Score | Coverage |
|---|---|
| 9–10 | Pillar exists + 8+ spokes + all interconnected |
| 7–8 | Pillar exists + 5–7 spokes, some gaps |
| 5–6 | Pillar exists + 3–4 spokes, significant gaps |
| 3–4 | No pillar OR only 1–2 spokes |
| 1–2 | Single thin page, no cluster |
Step 4 — Identify Missing Spokes
For each cluster, list every missing spoke type. Flag missing spokes as:
- Critical missing — commonly searched, competitor has it, we don't
- High value missing — good search volume, fits ICP
- Nice to have — lower volume but completes the cluster
- Skip — not relevant to [Your Brand]'s ICP
Step 5 — LLM Citation Gap
For each cluster, assess LLM coverage:
- Does any LLM currently cite [Your Brand] for queries in this cluster? (Use data from Task 40 — LLM Citation Gap Analysis if available)
- Which spokes are most likely to earn LLM citations? (Definition pages, comparison pages, data-rich how-tos > opinion pieces)
Step 6 — Internal Linking Audit (per cluster)
For each cluster, check:
- Does the pillar link to all spokes? (Missing links = equity leak)
- Does each spoke link back to the pillar?
- Do spokes link to related spokes within the cluster?
- Are there orphan spokes (spokes with no links to them)?
Output: link gaps as input for 2C (Internal Linking Strategist).
Output Format
Cluster Map
CLUSTER: [Topic Name]
Topical Authority Score: X/10
==============================================
PILLAR PAGE:
URL: [url or "MISSING"]
Status: [Exists / Weak / Missing]
Target keyword: [keyword]
Action: [None / Strengthen / Create]
EXISTING SPOKES ([N] pages):
[url] — [spoke type] — Links to pillar? [Y/N]
[url] — [spoke type] — Links to pillar? [Y/N]
MISSING SPOKES ([N] gaps):
[Critical] [spoke type] — suggested slug: /[slug] — Est. volume: [X]
[High] [spoke type] — suggested slug: /[slug] — Est. volume: [X]
[Nice to have] [spoke type] — suggested slug: /[slug]
LINKING GAPS:
[source url] → should link to → [target url] (anchor: "[text]")
LLM CITATION STATUS:
Currently cited for: [queries or "none found"]
Highest citation opportunity: [spoke type] — why: [reason]
Cluster Priority Table
| Cluster | Score | Missing Spokes | Priority | First Action |
|---|---|---|---|---|
| Mobile App Testing | 4/10 | 6 critical | HIGH | Build pillar page |
| Emulators | 7/10 | 2 high value | MEDIUM | Add VS spoke |
| AI Testing | 2/10 | 8 critical | HIGH | Build cluster from scratch |
Content Creation Priority List
Ordered by: impact on topical authority × search volume × effort
Priority 1: [page title] — Cluster: [X] — Type: [pillar/spoke] — Slug: /[slug]
Why first: [missing pillar / highest volume gap / competitor has it we don't]
Est. word count: [N] | Brief: run 2A on keyword "[keyword]"
Priority 2: ...
Internal Linking Action List (for 2C)
List all cluster linking gaps found in Step 6 in the format: Source URL | Target URL | Suggested anchor text
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.