- Home
- Skills
- Marketing & SEO
- Keyword-to-Page Planner for Small Sites
Keyword-to-Page Planner for Small Sites
Decide which existing or new page should own each keyword. Get a practical content map, separate product/guide/comparison/template/tool recommendations, potential overlap warnings, and an evidence-labeled priority queue.
$4
Works with the AI Agents you already use
Keyword-to-Page Planner for Small Sites
Example session with this skill installed
Fictional request with missing data
I run CedarWorks, a fictional planning app for small creative studios. I have three keywords: project brief template; how to plan a client project; project duration calculator. Plan for GB English, mobile, as of 2026-11-09. I have not supplied any search results or existing-page inventory. Give me a provisional plan and make the missing validation clear. Do not look anything up or publish.
All domains and business facts are synthetic demonstration inputs. The two JSON files separate normalization from agent hypotheses. The generated folder contains actual offline helper outputs.
- Read your context and instructions
- Compiled the keyword-to-page planner
- Generated the data export
Keyword-to-page plan
Planning date: 2026-11-09. Offline analysis of supplied data only. No live search results fetched or verified.
Coverage: 3 keyword rows; 3 clusters; 0 supplied pages; 0 supplied observations. Inventory status: unavailable. Potential conflicts: 0.
Evidence status
- hypothesis_no_sample: 3
Work queue
- P1 n-guide: validate_intent (planning score 0).
- P1 n-template: validate_intent (planning score 0).
- P1 n-tool: validate_intent (planning score -1).
Interpretation and limits
- Page formats and clusters are agent interpretations, not supplied facts. Inspect their rationales and evidence.csv.
- Sample support means at least 3 unique usable supplied results, a unique leading format with at least 60% share, and a match to the proposed format. It is not live verification or representative sampling.
- One latest eligible snapshot is selected per keyword, with sample_id as the same-date tiebreaker; snapshots are never pooled. Review conflicting samples before acting.
- Usability requires source/snapshot provenance, matching market/device and observation age at most 90 days; that cutoff is a review convention.
- Conflicts concern supplied declared targets/assignments. They do not establish actual ranking cannibalization. Missing inventory can hide conflicts.
- Search volume and result positions are supplied metrics only. Missing values remain blank. No traffic, competitor, ranking, or revenue figures were invented.
- Planning score = 2 × max business value + existing-page reuse bonus (1) − effort. Missing business value defaults to 1. It is an editorial sorting aid, not an SEO prediction.
- No pages, redirects, external accounts, or production settings were changed. A proposed tool/template still needs to be created and tested.
keyword-page.csv
keyword_id,keyword,cluster_id,task,page_type,owner_url,planned_action,evidence_status,evidence_ids,selected_sample_id,usable_sample_n,sample_leading_type,sample_leading_share,supplied_search_volume,volume_source,volume_as_of,volume_market,business_value,notes
n1,project brief template,n-template,Obtain a reusable project brief document,template,https://cedarworks.example/templates/project-brief,create,hypothesis_no_sample,,,0,,,,,,,1,business_value defaulted to 1; inventory unavailable; conflict/reuse checks incomplete
n2,how to plan a client project,n-guide,Learn a client-project planning process,guide,https://cedarworks.example/guides/client-project-planning,create,hypothesis_no_sample,,,0,,,,,,,1,business_value defaulted to 1; inventory unavailable; conflict/reuse checks incomplete
n3,project duration calculator,n-tool,Compute a project duration estimate,tool,https://cedarworks.example/tools/project-duration,create,hypothesis_no_sample,,,0,,,,,,,1,business_value defaulted to 1; inventory unavailable; conflict/reuse checks incomplete
conflicts.csv
conflict_id,type,keyword_ids,page_urls,cluster_ids,basis,next_step,priority
actions.csv
cluster_id,priority,planning_score,next_action,proposed_action,page_type,keyword_ids,effort,next_step,rationale
n-guide,P1,0,validate_intent,create,guide,n2,2,"Review the unsupported keyword rows; supply a dated, market/device-matched sample or revise the task/type.","Provisional explanation-task hypothesis, not observed SERP intent. Verify both intent and existing ownership."
n-template,P1,0,validate_intent,create,template,n1,2,"Review the unsupported keyword rows; supply a dated, market/device-matched sample or revise the task/type.",Provisional format hypothesis from the requested artifact. No SERP or inventory supplied; validate intent and check existing pages before creating it.
n-tool,P1,'-1,validate_intent,create,tool,n3,3,"Review the unsupported keyword rows; supply a dated, market/device-matched sample or revise the task/type.","The requested calculated result suggests a tool, but this is an unverified hypothesis. A real calculator and validation of its assumptions are separate build work."
evidence.csv
evidence_id,keyword_id,sample_id,source,url,title,snippet,supplied_rank,observed_at,market,device,agent_page_type,usable_in_sample,usability_note
evidence.csv
CSV · data export
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
Stop creating pages simply because two keywords look different. This Skill groups compatible search tasks, assigns a primary page owner, reuses suitable existing pages, and identifies when two declared targets or incompatible page assignments need review.
It distinguishes a downloadable template from an interactive tool, a guide from a product page, and a comparison task from a purchase task. When evidence is missing or mixed, the plan says so and gives you validation work instead of pretending the decision is confirmed.
What you get
- A reusable English Skill with setup guidance and clear input requirements.
keyword-page.csv: a keyword-to-owner map with format, evidence status, and supplied metric provenance.conflicts.csv: potential declared-target overlap and owner-format conflicts, with next checks.actions.csv: a dependency-aware work queue, using transparent business-value and effort assumptions.evidence.csvandsummary.md: a traceable supplied-data record and short planning summary.- An optional offline Python validator/exporter, 37 behavior tests, and two fictional worked examples with raw inputs and actual generated outputs.
Bring these inputs
Your site and offer, audience, market/device, keyword list, and existing-page inventory. Paste readable tables or attach CSV/other readable exports. Add any search-result snippets you are permitted to use, ideally with URLs, dates, market, and device. Search volumes are optional and must have provenance.
No SERP sample? You still get a hypothesis-led plan and validation queue. No page inventory? Reuse and overlap checks remain explicitly incomplete, and proposed new pages need an inventory check first.
How to use
Import the keyword-to-page-planner folder with your agent's supported Skill workflow. Ask:
Use Keyword-to-Page Planner to map these keywords and supplied SERP snippets to my existing pages. Identify unnecessary new pages, potential overlaps, and the first actions to take. Do not publish anything.
A text-capable agent performs semantic analysis. Python 3.10+ is optional for the included local validator and CSV exports; it uses only the standard library. Without file execution, the agent can produce the same planning tables and disclose that automated checks were not run.
Honest limits
This package does not fetch live search results, crawl your site, supply keyword volumes, verify rankings, prove SEO cannibalization, or forecast traffic/revenue. Supplied samples are not independent verification. It does not build tools, write every proposed page, publish, redirect, or modify your website.
The helper is a validator/exporter, not an automatic semantic classifier. It supports up to 300 keywords, 100 existing pages, and 3,000 supplied observations per run; larger projects need reconciled batches. English is the tested planning language. Examples are fictional demonstrations, not customer results.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Marketing & SEO.
No live SERP fetching, crawling, ranking measurement or traffic forecast. Supplied evidence is not independent verification. Helper is a validator/exporter, not a semantic classifier. Limits: 300 keywords, 100 existing pages, 3,000 observations per run.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 8 days ago
Nothing to install. It runs in your AI agent.
- Optional
- Python 3.10+ for CSV validation/export
Permissions
Uses only user-supplied keyword, page and permitted SERP files. Optional local Python validator and CSV output. Does not fetch search results, connect paid data APIs, or modify websites.