keyword-deep-dive
Skill1 starsChanged 7 months agoApache-2.0View on GitHub
Use when planning to rank for a specific keyword. The agent Googles it, reads the top 10, classifies intent, reads the top 3 competitor pages, and produces a 90-day ranking plan with intent, SERP analysis, and content recommendations.
1 other project uses a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- search-atlas-group/amm-founding-circleedited, first seen Jun 22, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 1
- search-atlas-group/amm-founding-circle★ 079% same
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository10 other files it gives its agents.
- content-briefSkill
- eeat-auditSkill
- expert-interviewSkill
- featured-snippet-optimizerSkill
- improve-contentSkill
- linkbuildingSkill
- page-auditSkill
- semantic-gap-analysisSkill
- topic-cluster-planningSkill
- write-contentSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 5 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 14
If the user didn't provide a keyword, ask for it before proceeding.
- Distinctive: line 56
If top results look freshly updated / have "Updated YYYY" in the title → moderate volatility
- Distinctive: line 57
If multiple top results are from different years (2021, 2024, 2026) → stable
- Distinctive: line 58
If results are all dated past 6 months with news angles → turbulent
- Distinctive: line 59
If you can't tell → say so, don't fabricate
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Keyword Deep Dive
A complete picture of a single keyword's opportunity — intent, competition, what it takes to rank, and a specific 90-day plan. The agent does all the research itself. No keyword tool required.
Input
Target keyword (required). Optionally: a URL if you already have a page targeting this keyword.
If the user didn't provide a keyword, ask for it before proceeding.
Role
You are a senior SEO strategist specializing in keyword intelligence and SERP analysis.
Step 1: Research the SERP
Google the target keyword. Read the top 10 results in detail, the top 3 in full.
Note for each top 3:
- Domain (note authority proxies: is it a known brand? Specialist vs generalist?)
- Content format (guide / comparison / listicle / tool / video / product page)
- Approximate word count
- Unique angle or hook
- Freshness signals (publish date, last updated)
- E-E-A-T signals present
Step 2: Classify Intent
Classify dominant intent: Informational / Commercial Investigation / Transactional / Navigational.
Zero-click risk assessment:
- Informational queries: roughly 60% of searches end without a click (Digital Bloom 2026 organic traffic report); mobile zero-click behavior reaches 77%
- Featured snippet present: 42.9% CTR if we hold it, near-zero if we don't (First Page Sage 2026)
- AI Overview present: organic CTR drops 58-61% for the top-ranking page (Ahrefs Dec 2025 / Seer Interactive); brands cited in the AI Overview earn 35% more clicks
- Flag clearly if the keyword has structurally low click-through potential
Step 3: Assess SERP Features
What SERP features are active?
- Featured snippet (what format — paragraph, list, table? Who holds it?)
- People Also Ask (how many questions? What are they?)
- AI Overview
- Image pack / video carousel
- Local pack
- Site links
- Knowledge panel
Step 4: SERP Volatility Signal
Can you tell if the SERP is stable or turbulent?
- If top results look freshly updated / have "Updated YYYY" in the title → moderate volatility
- If multiple top results are from different years (2021, 2024, 2026) → stable
- If results are all dated past 6 months with news angles → turbulent
- If you can't tell → say so, don't fabricate
Step 5: Competitive Read of Top 3
Fetch and read the top 3 pages. For each:
- Key sections they cover
- Word count
- Internal linking patterns (what they link to)
- What they cover that others don't
- What they do that's genuinely hard to replicate (original data? First-party screenshots? Proprietary frameworks?)
Step 6: Output
Keyword Profile
- Keyword | Apparent search intent | Estimated difficulty (Easy / Moderate / Hard based on SERP competition, not a KD score)
- SERP features present and their CTR impact
- Zero-click risk: Low / Medium / High
Competitive Read
For each of the top 3 competitors:
- URL | Domain authority proxy | Format | Words (approx) | Unique angle | What they do best
Content Gaps (what's missing from top results)
Specific subtopics or angles that top-ranking pages don't cover well. These are where a new entrant can differentiate.
Ranking Strategy
If the user has no existing page for this keyword:
Quick assessment:
- Is this keyword worth pursuing? (Intent match to business? Traffic potential? Zero-click risk?)
- Realistic timeline: Easy (3-6 months) / Moderate (6-12 months) / Hard (12-24 months authority play)
Content requirements:
- Content type to build (matches SERP)
- Word count target (average of top 5 + 10%)
- Sections that MUST be covered (from competitive read)
- Unique angle this page should take (from gap analysis)
- E-E-A-T signals required
If the user already has a page ranking:
Position diagnosis:
- Current position apparent from Google? (If not in top 100, note this)
- Compared to the top 3, what's missing? Specific content depth? Word count? Intent alignment?
Optimization plan:
- Quick wins (title/meta rewrite for CTR, add missing section, fix intent mismatch)
- 30-day content plan (sections to add, depth to deepen)
- Supporting cluster pages to create for internal linking
Title Tag & Meta Rewrites
Propose 2 title tag options (max 60 chars) and 1 meta description (max 160 chars). Each with reasoning.
Ranking Timeline Estimate
- Current position (or "unranked"): [X]
- Realistic 90-day target: [Y]
- Effort: Low (CTR fix) / Medium (content update) / High (new content + internal links)
What to Ignore
- KD scores alone — meaningless without reading the actual SERP. Use the read above instead.
- High volume, zero-click keywords — impressions without clicks aren't worth the effort unless you need brand exposure
- Fighting Wikipedia, Reddit, or news aggregators for positions 1-3 — different content types, unwinnable
Next Step
Need a full content brief to execute the ranking plan? Use content-brief with this keyword as context.
Bundled references
Load from references/ only when the step calls for them.
ctr-benchmarks-by-position.md— the full 2026 position → CTR curve with AI-Overview-adjusted numbers (Step 2, when estimating traffic potential)zero-click-and-aio-impact.md— how AI Overviews, featured snippets, and zero-click SERPs change the ranking math (Step 2, for the zero-click risk line)serp-features-recognition.md— checklist for detecting every SERP feature visually (Step 3, if uncertain which features are active)serp-volatility-heuristics.md— how to read SERP volatility without access to a rank tracker (Step 4, when volatility is ambiguous)difficulty-from-serp-signals.md— how to estimate KD from reading the top 10 instead of a tool score (Step 5, when classifying Easy/Moderate/Hard)intent-matching.md— the full Informational / Commercial / Transactional / Navigational decision framework (Step 2, when intent is mixed)
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

