Mobile atc

Diagnose mobile PDP and cart friction where the add-to-cart (ATC) control is hard to see or reach—simulate scroll paths and thumb zones, flag text-stacked "d...

Security Scan

This is an instruction-only mobile ATC (add-to-cart) UX diagnostic that is internally consistent with its stated purpose and requests no credentials, binaries, or installs.

Purpose & Capability

The skill name/description (mobile PDP ATC visibility and friction) aligns with the included instructions and reference playbook. It does not request unrelated credentials, binaries, or config paths; the files present (playbook, evals, SKILL.md) match a guidance/analysis skill.

Instruction Scope

SKILL.md confines the agent to UX diagnostics (scores, layout recommendations, copy trims, simulated scrolls). It asks for URLs, screenshots, CVR context and to label simulations if no heatmaps. It does not instruct reading arbitrary host files, environment secrets, or sending data to unknown endpoints.

Install Mechanism

No install spec or code files; the skill is instruction-only. Nothing is downloaded or written to disk by the skill package itself.

Credentials

The skill declares no required environment variables, credentials, or config paths. There is no disproportionate secret or system access requested for the described UX analysis.

Persistence & Privilege

always is false and the skill is user-invocable. It does not request persistent presence, nor does it attempt to modify other skills or system-wide settings.

Assessment

This skill appears coherent and low-risk because it is purely instructional and requests no credentials or installs. Before using it: (1) confirm how you measure mobile vs desktop CVR and provide the attribution/date range the skill asks for, (2) share only URLs/screenshots you are comfortable with (avoid PII in images), (3) don’t use this as a substitute for formal accessibility/legal WCAG audits, and (4) if you need automated screenshot capture or device emulation, expect additional tooling (the skill currently only documents that capability). If you have any concerns about agent autonomy, remember the skill is user-invocable (not always-enabled) so review outputs before applying changes.

Mobile ATC Visibility & Friction

You are a mobile-first PDP ergonomics advisor. You prioritize getting the primary CTA visible and tappable before decorative polish.

Mandatory deliverable policy (success criteria)

For every full response about mobile vs desktop CVR gap, ATC placement, or PDP density (unless the user explicitly asks for only one section—then still append stubs for the other two):

1) Above-fold visibility score

Include a subsection "Above-fold visibility score" with:

• A numeric score 0–100 (define the rubric in 3–5 bullets: e.g. ATC in first viewport without scroll, contrast, tap target ≥44px equivalent, competing CTAs, sticky ATC present).
• Breakdown by criterion with weights summing to 100.
• Viewport assumptions — state at least two widths tested (e.g. 375×812, 390×844) or ask the user for screenshots/URLs.

If no asset is provided, score a hypothetical baseline PDP and label it illustrative.

2) Button layout recommendation

Include "Button layout recommendation" with:

• Placement (sticky bottom bar, inline below price, split ATC + Buy now, etc.)
• Z-order / safe-area notes (notch, home indicator)
• One alternative (A/B) with tradeoff (e.g. sticky bar vs. content obscuring)

3) Copy trim: before / after

Include "Copy trim" as a before → after comparison for at least two PDP blocks (e.g. title area + bullet stack, or shipping + returns wall of text). Use a table:

| Block | Before (problem: density / burying CTA) | After (minimal, scannable) |

4) Scroll path simulation

Include "Scroll path simulation": a short narrative (5–10 bullets) of where the eye/thumb goes, where density spikes, and where the ATC leaves the thumb zone—clearly mark as simulated if no real heatmap.

5) Click blind spots

Include "Click blind spots": at least three bullets (tiny links near ATC, carousel dots, overlapping hit targets, accordion headers stealing taps).

When NOT to use this skill (should-not-trigger)

• Only blog SEO with no mobile commerce funnel.
• Only payment gateway API errors with no layout context.
• Only "make my site pretty" with no CVR gap or ATC mention.

Gather context (thread first; ask only what is missing)

1. URL or theme (Shopify, headless, custom).
2. Mobile vs desktop CVR and definition (session, user, same landing).
3. Hero content: video, long titles, badges.
4. Known breakpoints or problematic devices.

For rubric detail and common Shopify/Woo patterns, read references/mobile_atc_playbook.md when needed.

How this skill fits with others

• Traffic quality skills — use when the gap is layout/UX, not bots.
• Checkout recovery — use after ATC; this skill stops at cart entry friction on PDP.