Application plan

A mobile-first participatory-surveillance + early-warning app for Arizona, spanning two verticals (Vector-Borne Disease and Heat), built on the existing DuckLake knowledge graph and driven by an agentic architecture that consumes real-time data via MCP servers.

The five documents

0Overview & roadmap summary

The thesis in one paragraph, the architecture in one diagram, and the status snapshot of every component.

1Parameter mapping

How Figure 2's Minimum Set of Key Data Parameters maps to each vertical, with full class-by-class tables and the vertical-specific additions (heat: AC access, wearable skin-temp; VBD: bite location, attached duration).

2MCP integration

The nine-MCP server inventory, a parameter-by-parameter source map, and the multi-MCP joins behind "where can I cool off?" and VBD triage.

3Agentic architecture

The eight-agent topology — Intake → Geo-Enrichment → Validation → Triage → Enrichment → Notification, plus background Cluster Detection and Knowledge Update — and why agents and not a monolith.

4End-to-end data flows

Four worked scenarios: hiker mails in a tick; symptomatic patient calls 211; CHW-mediated unsheltered heat check-in; agency-side cluster review.

5Roadmap

Five phases from hackathon MVP to a 12-month statewide pilot, each tied to a success criterion against the Figure 3 timeliness milestones. Risk register included.

6One Health reporting app

Anonymous-first intake for Human / Animal / Environmental events — low-text UI, EXIF-stripped photos, ZIP/km coarsening, never-diagnose risk-assessment boundary, DuckLake snapshots backed by GitHub LFS. Web pilot → iOS + Android.

7Authentication

Optional account on top of anonymous-first reporting. Email + password, magic-link, and OAuth (Apple, Google, Facebook — Apple first per App Store Guideline 4.8). Supabase Auth backend; right-to-erasure detaches but does not delete attached reports.

Architecture at a glance

   ┌──────────────────────────────────────────────────────────────────┐
   │                       Community user (mobile)                   │
   │      VBD UI · Heat UI · SMS fallback · Voice (transcribed)      │
   └─────────────────────────────────┬───────────────────────────────┘
                                     │  free-text / photo / geo / form
                                     ▼
   ┌──────────────────────────────────────────────────────────────────┐
   │                           Agentic layer                          │
   │  Intake → Geo-Enrichment → Validation → Triage → Enrichment      │
   │                            ↓                                     │
   │       Notification ← Cluster Detection ← Knowledge Update        │
   └─────────────┬──────────────────────────┬─────────────────────────┘
                 │                          │
                 ▼                          ▼
   ┌──────────────────────┐     ┌───────────────────────────────────┐
   │  DuckLake knowledge  │     │           MCP servers              │
   │  graph (this repo)   │ ←→  │  vectorsurv-mcp ✅                 │
   │                      │     │  nws-heatrisk-mcp ⏳               │
   │  observations,       │     │  mag-hrn-mcp ⏳                    │
   │  pathogens,          │     │  adhs-mcp ⏳                       │
   │  outbreaks,          │     │  great-az-tick-check-mcp ⏳        │
   │  resources, regions, │     │  whispers-mcp ⏳                   │
   │  tribes, ICD-10, …   │     │  inaturalist-mcp ⏳                │
   │                      │     │  211-az-mcp ⏳                     │
   └──────────────────────┘     │  knowledge-graph-mcp ⏳            │
                                └───────────────────────────────────┘

The thesis in one paragraph

The Minimum Dataset (Figure 2) is the shared data contract that lets one app serve both verticals: the General class is identical for VBD and Heat, the Human class diverges into infectious symptoms vs. heat-illness symptoms, the Exposure and Environmental classes pick up domain-specific sub-fields, and the Livestock / Wildlife classes matter for VBD but are silent for Heat. Treating the dataset as a contract — rather than building two apps — means one knowledge graph, one set of MCP integrations, one agentic pipeline, and two thin domain-specific UIs on top.

How this connects back

This plan referencesWhere
Figure 2 — Minimum Set of Key Data Parametersfigures/index.html#fig-02
Figure 3 — Outbreak Timeliness Milestonesfigures/index.html#fig-03
Figure 4 — 12-step design lifecyclefigures/index.html#fig-04
Wildlife & Vector-Borne Diseases focus groupwildlife/
Heat focus groupheat/
vectorsurv-mcp (the only MCP server shipped today)mcp/vectorsurv-mcp/
Pathogen knowledge graphgraph/
Arizona map (geospatial slice)map/
DuckLake schema seedsschema/ · schema/deep/