agentleFS
Aetheria-Labs1Aetheria-Labs1/storefront-skills/skills/plan-page/SKILL.md

plan-page

Skill8 starsChanged 3 months agoMITView on GitHub

Plan one storefront page from strategy through final copy, section order, proof, offers, template direction, responsive layout, and production asset requirements. Stops at an approved execution-ready specification.

4 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions4 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 4

    1. udecode/plate★ 17ksame name
    2. udecode/dotai★ 1.2ksame name
    3. udecode/better-convex★ 452same name
    4. Pablomg02/pamaga-agent-toolkit★ 1same name

    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 repository12 other files it gives its agents.

The file

Show its rules and text

We split this skill into the rules it gives an agent

16 rules. 0 are common, 10 are distinctive, and 6 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 10

    Distinctive

    This author's own choices. Worth a look.

  • 6

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If the request is empty or incomplete, ask every material question needed to make these decisions.

    line 72
  2. Very specific:

    Stop with PLAN_BLOCKED when a missing choice or evidence would materially change the page.

    line 77
  3. Very specific:

    Read review availability before planning proof.

    line 86
  4. Distinctive:

    Use the guide's default anatomy as the starting point.

    line 119
  5. Distinctive:

    Use the Consumer decision model to list the shopper's top decision questions, concerns, objections, proof needs, and mobile context.

    line 125
  6. Distinctive:

    place proof beside the claim it supports.

    line 132
  7. Distinctive:

    Use the Page strategy rationale to place component explanations, mechanism evidence, major objections, and value-supporting imagery in that …

    line 136
  8. Distinctive:

    Write the final customer-facing content for every section:

    line 145
  9. Very specific:

    No placeholders, writing instructions, draft options, or TBD values may remain.

    line 154
  10. Distinctive:

    Search templates only after the section requirements are complete.

    line 186
  11. Distinctive:

    Search page kits for the selected type, vertical, mood, and objective.

    line 188
  12. Distinctive:

    When selection UI is available, let the user select and wait for the selection response.

    line 189
  13. Distinctive:

    If no page kit fits, search sections only for the specific planned sections that need a reusable direction.

    line 195
  14. Very specific:

    If the shelf is empty or every candidate conflicts with the plan, record a custom layout direction.

    line 197
  15. Very specific:

    Write every block in references/plan-page.md in its defined order.

    line 205
  16. Very specific:

    When the page sells an offer, read references/offers/offer-stacking.md and record the stack in the plan: blockers, the item answering each, …

    line 233

“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Plan a Storefront Page

Produce the complete specification that later skills visualize, resource, and design. Finalize the strategy and customer-facing copy here. Do not choose islands, author code, create drafts, or produce assets.

Read Only What Applies

Always read:

  • work/storefront/setup.md;
  • the selected store's brand.md, design.md, products.md, persona.md, rules.md, and default theme CSS;
  • references/plan-page.md;
  • references/page-type-guide.md;
  • references/planning-rules.md.

These three references are the base planning packet. General setup context is a map, not live commerce authority.

For selling pages, including full-price products with no promotion, also read references/offers/offer-strategy.md before choosing the page type or sections. For informational or support pages, record applicability without forcing an offer.

Use:

  • lexsis_catalog.list and lexsis_catalog.get to refresh selected products;
  • lexsis_catalog.reviews_status, lexsis_catalog.review_collections, lexsis_catalog.reviews, and lexsis_catalog.reviews_search for proof;
  • lexsis_template_library.search_page_kits, lexsis_template_library.search_sections, and lexsis_template_library.get_kit after sections are planned.

Product finders, assessments, branching questions, and guided recommendations are supported by the live Quiz island. For a quiz plan, read references/quiz-authoring.md and route implementation to /quiz. Plan the entry, question paths, deterministic results, real products/variants, and exit or cart actions. If saving is needed, include purpose, disclosure, question classification, visibility, retention, resume window, and separate contact choice. Keep response saving independent from analytics consent.

Games, reward issuance, or custom interactions still require a dedicated live island schema; do not invent a runtime or hidden controls. Do not promise automated result email, preview capture, or an unshipped quiz funnel dashboard.

Resolve unfamiliar schemas through exact router/action discovery. Use authoritative web research for public facts and ask the user about private merchant decisions, unclear positioning, intended campaign messaging, or conflicting evidence. Record the source and date for researched facts.

1. Establish the Brief

Identify:

  • workspace, store, and default working theme;
  • requested page and page type;
  • product, collection, or whole-store scope, and whether the layout is one page or a product or collection template that serves many handles (references/resource-templates.md; record the sample handles);
  • target market and persona;
  • traffic source and supplied campaign message;
  • funnel stage and awareness;
  • conversion goal and CTA destination;
  • offer, purchase model, dates, and regions;
  • proof, claims, legal constraints, and mandatory sections.

If the request is empty or incomplete, ask every material question needed to make these decisions. There is no fixed question limit. Group related questions so the user can answer efficiently. Do not ask for facts already available in setup, the catalog, or reliable public sources.

Stop with PLAN_BLOCKED when a missing choice or evidence would materially change the page. Lesser uncertainties remain labelled in ## Open decisions.

2. Refresh Live Context

Find likely products in products.md, then refresh every selected product with the live catalog. Capture exact ids, variants, options, prices, availability, selling plans, descriptions, and product-media jobs.

Read review availability before planning proof. For a PDP or product template, ask the user what a product with no reviews shows: store reviews (default), one review collection from lexsis_catalog.review_collections (only active ones render), or none; record it as the plan's review fallback (references/resource-templates.md, "Reviews on product pages"). Verify public policy, occasion, certification, shipping, returns, and claim facts from authoritative sources when they are not available through Lexsis. Merchant-controlled conflicts return to the user.

For a supplied campaign message or creative, extract its promise, nouns, CTA, tone, claims, and visual cues from the material the user supplied. Do not invent a campaign context.

Develop the Offer Value Strategy

For selling pages, follow references/offers/offer-strategy.md to connect customer evidence to the desired outcome, differentiation, mechanism, complete solution, value, proof, and one testable uncertainty. Consider value improvements before proposing discounts. Complete the seven value-strategy fields in Page strategy before finalizing sections and copy; attach section IDs once the narrative is chosen.

Respect fixed merchant offers and supplied campaign promises. Keep unconfirmed prices, bonuses, or alternative positioning in Open decisions and out of final copy. Missing economics limits profitability conclusions without blocking a valid existing offer. These decisions do not authorize commercial changes.

3. Identify One Page Type

Walk references/page-type-guide.md before template search. Select exactly one type using traffic, stage, awareness, desired action, product count, offer, and campaign trigger.

Use the guide's default anatomy as the starting point. Record every mandatory section omitted and every deviation with a reason. Review the checklist before presenting the plan.

4. Build the Consumer Narrative

Use the Consumer decision model to list the shopper's top decision questions, concerns, objections, proof needs, and mobile context. Turn them into an ordered section narrative:

  1. establish relevance and message match;
  2. explain the product or mechanism;
  3. resolve the largest objections;
  4. place proof beside the claim it supports;
  5. present the offer and risk reversal;
  6. close with one clear action.

Use the Page strategy rationale to place component explanations, mechanism evidence, major objections, and value-supporting imagery in that narrative. The strategy fields are not additional mandatory sections.

Sections are decided before templates. Each section must have a distinct consumer job; merge or remove sections that repeat the same job.

5. Finalize Copy, Proof, and Layout

Write the final customer-facing content for every section:

  • eyebrow, headline, subhead, body, labels, bullets, CTA, and destination;
  • FAQ questions and complete answers where applicable;
  • exact offer terms and policy language;
  • proof and claim ids;
  • desktop and mobile layout intent;
  • interaction intent described as shopper behaviour, never implementation.

No placeholders, writing instructions, draft options, or TBD values may remain. Every number, timeframe, quote, badge, certification, price, or guarantee must resolve through the Proof ledger, Offer ledger, or Claim gate.

6. Define Production Asset Requirements

Define what each section needs; leave sourcing and production to /plan-assets.

For every slot specify:

  • section and consumer job;
  • product or variant identity;
  • real-only, editable-real, or generation-permitted status;
  • mobile aspect ratio, crop, focal point, and minimum pixel dimensions first;
  • larger-screen aspect ratio, crop, minimum pixel dimensions, and whether it derives from the same master;
  • subject, composition, focal point, quiet zone, palette, and neighbouring section fit;
  • evidence, rights, or provenance requirements;
  • optional concept-frame reference when /visualize-page is later used.

Requirements may cite real Shopify media already identified through the catalog, but they do not select library assets, upload files, import URLs, generate images, spend credits, or declare production assets ready.

Official marks, certifications, press logos, customer proof, results imagery, packaging identity, and exact product appearance always require genuine authorized media.

7. Choose Template Direction

Search templates only after the section requirements are complete.

  1. Search page kits for the selected type, vertical, mood, and objective.
  2. When selection UI is available, let the user select and wait for the selection response.
  3. Without selection UI, follow the user's delegation:
    • present the strongest candidates when they want to choose;
    • choose the best fit and explain why when they delegate the choice.
  4. Resolve a selected kit with lexsis_template_library.get_kit.
  5. If no page kit fits, search sections only for the specific planned sections that need a reusable direction.
  6. If the shelf is empty or every candidate conflicts with the plan, record a custom layout direction. /design-page will author it.

Template direction never changes the consumer narrative merely to fit an available template.

8. Produce and Approve the Plan

Write every block in references/plan-page.md in its defined order. The result contains:

  • complete strategy and consumer decisions;
  • final copy;
  • ordered sections and responsive layout;
  • proof, offer, message-match, and claim decisions;
  • template or custom direction;
  • production asset requirements;
  • open decisions and status.

Present a compact summary with page type, audience, goal, offer, section count, asset-requirement count, proof/claim risks, and template direction. Revise until the user approves.

Status rules:

  • PLAN_BLOCKED: a material strategy, copy, offer, proof, claim, product, or CTA decision is unresolved;
  • PLAN_READY_FOR_APPROVAL: the specification is complete and awaits the user's explicit approval;
  • PLAN_APPROVED <who> <date>: the user explicitly approved this exact plan.

Plan approval does not authorize asset spending, draft creation, or publication.

Offer Stack

When the page sells an offer, read references/offers/offer-stacking.md and record the stack in the plan: blockers, the item answering each, real values, guarantee, and limits. Plan headers and footers as shared sections from setup.md; do not write new copy for them.

Return

Return the complete plan, binding, status, and next route:

  • optional /visualize-page when the user wants concept approval;
  • /plan-assets to resolve production media;
  • back to /plan-page whenever a visual or asset decision changes strategy, sections, claims, or final copy.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.