skills.rest
chriso789
chriso789COMMUNITY
Explore all skills by this author →
GitHub

measurement-overlay-visual-qa

Validates aerial-backed measurement overlay rendering and manual perimeter verification UI.

Updated Sep 24, 2025
Search Tags:#react#manual verification#visual qa#roof measurement#aerial imagery#perimeter overlay#debug viewer
One-click install
npx skills add https://github.com/chriso789/pitch-1 --skill measurement-overlay-visual-qa-chriso789
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: measurement-overlay-visual-qa
Source: https://github.com/chriso789/pitch-1/tree/main/.agents/skills/measurement-overlay-visual-qa
Command: npx skills add https://github.com/chriso789/pitch-1 --skill measurement-overlay-visual-qa-chriso789

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI roof measurement runs can silently produce blank reports, geometry-only diagrams, or premature customer-ready exports. This Skill enforces a consistent visual QA surface so a human can verify perimeters, masks, rejected regions, and roof-line diagnostics against real aerial imagery before any export. ## Core Features & Use Cases - Aerial-first overlay enforcement: Requires the aerial raster as the background with 11 independently toggleable diagnostic layers (raw/refined/selected perimeter, masks, rejected regions, DSM ridge/valley/hip candidates). - Blocked-topology fallback: Guarantees failed topology runs still render the perimeter debug overlay instead of a blank report card. - Manual perimeter governance: Defines the manual editor toolset (drag vertex, add corner, snap-to-edge, rerun) and prevents a manually verified perimeter from flipping customer_report_ready without downstream gates. - Use Case: When updating MeasurementReportDialog or the AI Process Viewer, invoke this Skill to produce the component list, layer contract, JSON field mapping, and acceptance checklist before writing code. ## Quick Start Audit the MeasurementReportDialog and overlay components against the visual QA rules and list every missing layer, metric, and gate violation.

Frequently Asked Questions about measurement-overlay-visual-qa

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I verify an AI roof measurement perimeter visually?▼

Render the aerial raster as the background and overlay the raw, refined, and selected perimeters plus the target mask and rejected regions as toggleable layers. The reviewer then uses Approve, Edit, or Reject controls backed by metrics like visual_edge_alignment_score and corner_snap_confidence.

What overlay layers should a roof measurement debug view include?▼

Eleven layers: aerial background, raw perimeter (gray), refined perimeter (green), selected perimeter (blue), target mask fill, global building mask (dashed), unsupported segments (red), corner cuts (orange), rejected regions (hatched), corner snap markers, and DSM ridge/valley/hip candidates.

What should render when roof topology detection fails?▼

A blocked-topology run must render the perimeter debug overlay in place of the report, never a blank card. States like ai_failed_perimeter or perimeter_only still show the aerial-backed perimeter diagnostics.

Does a manually verified perimeter make the report customer-ready?▼

No. A manual perimeter only unlocks topology diagnostics. It sets perimeter_source_locked to user_verified_perimeter and holds customer_report_ready false until topology, typed roof_lines, valid pitch, and vendor benchmark gates pass.

Why does the overlay show a geometry-only SVG instead of aerial imagery?▼

That happens when no aerial source (raster_url, Google Static Maps tile, or persisted DSM-aligned RGB) is available. Geometry-only is only acceptable as a fallback and must display a visible no-aerial-available banner.

Related Skills

Explore More Software Engineering Skills→
zjunlpzjunlp
1.1k

scienceworld-measurement-taker

Measure quantitative properties of substances using designated tools.

Official
Intermediate
t0ddc3byt0ddc3by
58

outcome-statement-builder

Transform capability inputs into verifiable outcome statements with a two-phase approval gate.

Community
Advanced
Open-PawsOpen-Paws
3

code-review

Automate layered code review for pull requests in CI pipelines.

Official
Advanced
WebWakaHubWebWakaHub
20

API Tester

Validate APIs for functional, security, and performance requirements.

Official
Advanced