Low-Visibility UI Audit → Handoff Docs#

Audit existing mobile UI for field low-visibility conditions — environmental (low light, glare) + physical (gloves, wet hands, vibration) — and emit a structured handoff document set for another agent to implement. Field conditions only; disability/a11y populations are out of scope — redirect them.

This skill never edits the target. Its stable, must-always-succeed deliverable is the doc set ( audit.json + report.md), written only to the out dir. It is built to run many rounds on a scoped subset — keep each run bounded (targeted pages/components, capped scan).

Protocol — analyze, document, hand off (never edit the target)#

1. Scope — confirm a field low-visibility audit of an EXISTING mobile UI; else redirect (new design / a11y / generic CSS = out of scope). Read the targeted request ( --pages, --selector); with none, a bounded default scan.
2. Preflight — locate the in-scope files for the targeted scope only: H5 .html/ .css, or mini-program .wxml/ .wxss. Pick the visual-pass render method ( rules/visual-pass.md); if none is available, degrade — still emit docs.
3. Analyze — python3 scripts/audit.py <target> [--pages a,b] [--selector .x] [--css sheet.wxss]. Deterministic findings thresholded by design-tokens.json.
4. Visual pass (bounded) — render the targeted page(s) at a mobile viewport, resolve the needs_judgment long tail by visual reasoning, label visual findings as estimates. Never fabricate a ratio.
5. Synthesize — audit.py writes the doc set to the out dir (default <target>/.lv-audit/). Confirm audit.json validates and every section is present, including the zero-findings "clean" doc.
6. Report — summarize findings by severity, point at the doc paths + residual needs_judgment, suggest the next-round scope. Do not edit the target.

Output goes only to the out dir (writing into the target source tree → OUTPUT_ONLY refusal). No network beyond rendering the target. Refuse to recommend lowering contrast/target below the field threshold for a field-critical control; offer a compliant alternative. Severity, interpretation, doc-set spec, and degraded mode → the modules below.

Modules & scripts#

PathWhenrules/audit-protocol.mdStep 3–4: severity model, reading findings, completing needs_judgment, degraded mode, adversarial holdsrules/handoff-docs.mdStep 5: doc-set sections + how fixes are written as recommendations (not applied)rules/visual-pass.mdStep 2/4: per-target render method, what the visual pass catches, the page cap, graceful degradationreferences/design-tokens.jsonthreshold values (single source of truth)references/field-conditions.mdwhy each threshold exists; standards citationsassets/handoff-doc.template.mdthe doc-set shape an implementer agent readsassets/fix-snippets.htmlcompliant component patterns referenced by recommendations (never applied)scripts/audit.py <target> [--pages a,b] [--selector .x] [--css f] [--out d] [--input-mode static|visual_estimate] [--json]scope → analyze → emit doc set; never edits the target (exit 3 = empty scope)scripts/analyze.py <file> [--css f] [--selector .x] [--viewport-px n]deterministic findings JSON for one file (H5 or WXML+WXSS)evals/run_all.pyregression + behavioral harness — every adversarial edge bound to a case