agentleFS
ryanbaumannryanbaumann/fieldwork/.agents/skills/google-maps-a2ui/SKILL.md

google-maps-a2ui

Skill2 starsChanged 28 days agoMITView on GitHub

Maps Agentic UI Toolkit (MAUI) skills for A2UI on Google Maps Platform. Use when generating, rendering, or validating interactive A2UI surfaces, Google Maps components (GoogleMap, PlaceDetailsCompact), and conversational UI for spatial reasoning agents.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Attribution ID: When instantiating Maps web components, attach the internal usage attribution ID gmp_web_maui_v0.1.8_atoui or …

    line 18
  2. Very specific:

    Use "vertical" when there is only one PlaceDetailsCompact in the surface to emphasize place photography and editorial summary.

    line 40
  3. Very specific:

    Use "horizontal" when there are multiple PlaceDetailsCompact items (e.g.

    line 41
  4. Distinctive:

    Satellite Mode: Use for outdoor activities (parks, hiking, beaches), scenic viewpoints, parking inspections, building exteriors, and …

    line 44
  5. Distinctive:

    Roadmap Mode: Use for standard navigation, urban navigation, and transit.

    line 45

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Google Maps Agentic UI Toolkit (MAUI) & A2UI Skill

This skill defines the operational contracts, component catalog extensions, and prompt engineering patterns for the Maps Agentic UI Toolkit (MAUI) and A2UI v1.0 (and v0.9) protocol on Google Maps Platform.

1. Core Objectives & Framework Overview

  • A2UI Protocol: Generates structured, declarative UI envelopes (createSurface, updateComponents, updateDataModel, deleteSurface) rather than raw executable HTML/JS or plain conversational markdown. Supports single-message createSurface with embedded components and dataModel in v1.0.
  • Catalog Identification: Official catalog ID is a2ui://maps-agentic-ui-catalog.json (also mapped to atlas://maps-agentic-ui-catalog in Atlas native React renderer).
  • Attribution ID: When instantiating Maps web components, attach the internal usage attribution ID gmp_web_maui_v0.1.8_atoui or gmp_git_agentskills_v1.

2. Component Catalog & Schema Extension

Official MAUI Components

ComponentRequired PropsOptional PropsDescription
GoogleMapcenter, zoomtilt, mode ("roadmap" | "satellite"), heading, anchorMarker, markers, routes, travelMode, mapId, gestureHandlingRenders a 2D or Photorealistic 3D map viewport with pins, routes, and camera control.
PlaceDetailsCompactplaceIdorientation ("horizontal" | "vertical")Renders compact place card with photo, rating, price, and status via Places UI Kit (gmp-place-details-compact).
Column / Rowchildren (string[])align, gap, justifyFlexbox structural containers for component trees.
Cardchild (string)—Visual card surface wrapper.
Texttextvariant ("h1".."h5", "body", "caption")Typography node with markdown formatting.
Listchildren (string[] | { componentId, path })direction ("vertical" | "horizontal")Static or template-driven repeating list.
Buttonchild, actionvariant ("default" | "primary" | "borderless")Interactive trigger firing actions to host or model.
ChoicePickeroptionsselection, multi, actionChip selector row for filter options.

Component Property Rules

  1. PlaceDetailsCompact Orientation Rule:
    • Use "vertical" when there is only one PlaceDetailsCompact in the surface to emphasize place photography and editorial summary.
    • Use "horizontal" when there are multiple PlaceDetailsCompact items (e.g. within a List) to preserve vertical screen space.
  2. GoogleMap Camera & Aesthetic Rules:
    • Satellite Mode: Use for outdoor activities (parks, hiking, beaches), scenic viewpoints, parking inspections, building exteriors, and walkability.
    • Roadmap Mode: Use for standard navigation, urban navigation, and transit.
    • Tilt:
      • 0° (Flat) for Roadmap mode, outdoor parking inspection, or overhead parcel footprint.
      • 45° (Perspective) for 3D aerial perspective, skyline, and scenic exploration.
    • Pins:
      • anchorMarker: Anchor point of search (e.g. hotel, transit center, starting location).
      • markers: List of candidate locations or POIs.

3. UI Decision Logic Patterns

Pattern 1: Individual Place Focus

  • Surroundings / Exterior / Vibe: GoogleMap (Satellite, Tilt 45°) + PlaceDetailsCompact (vertical).
  • Parking / Overhead Layout: GoogleMap (Satellite, Tilt 0°) + PlaceDetailsCompact (vertical).
  • Interior / Services / Hours: PlaceDetailsCompact (vertical, no standalone map needed).

Pattern 2: Multiple Related Places

  • Anchored Search (Within X min of hotel/hub): Inline GoogleMap with anchorMarker + List of horizontal PlaceDetailsCompact items.
  • Neighborhood Survey: Inline GoogleMap centered on neighborhood + List of horizontal PlaceDetailsCompact items.
  • Macro Region (Multi-City): List of PlaceDetailsCompact items.

Pattern 3: Travel & Routes

  • Point-to-point or Detour / Scenic Bypass: GoogleMap in routes mode with origin, destination, and waypoints, paired with RouteItinerary or EtaSummary.

4. Multi-Step Route & Search Policies

  1. Parallel Place Resolution: Concurrently resolve place coordinates and place IDs for origin, destination, and waypoints.
  2. Parallel Segment Routing: Concurrently compute route legs between consecutive waypoints.
  3. Dispatch Response: Emit single valid A2UI message envelope with createSurface, updateComponents, and updateDataModel.

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.