CatchIQ

A fishing catch tracking app. Log catches with species, bait, weather, and location data, then use AI analytics over your catch history to find patterns and catch more fish.

Features

• Dashboard — totals, personal bests by species, best spot, recent catches, and a saved-spots map at a glance
• Catch logging — species, bait, and spot pickers; location comes from the device via the browser Geolocation API, and the weather at that moment (air temp, wind, pressure, sky) is auto-filled from Open-Meteo
• AI Insights — Claude analyzes the full catch log and writes personalized best time of day, best bait, and best conditions breakdowns plus a summary
• History - searchable catch history

Planned Features

• Spots and Map — save and edit custom spots, and an interactive Leaflet map

Tech stack

Backend: .NET 10, ASP.NET Core Web API, EF Core 10, SQL Server, ASP.NET Identity + JWT, Swashbuckle, Anthropic C# SDK

Frontend: React 19, TypeScript, Vite 8, Tailwind 4, React Router 7, React Leaflet, NSwag

External services: Open-Meteo (weather at catch time), Claude (insight generation), OpenStreetMap tiles, browser Geolocation API

Tooling: GitHub Actions CI

Architecture

Controller → Manager → Engine → Accessor

Layer	Responsibility
Controller	HTTP only: routing, model binding, status codes
Manager	Business rules, validation, orchestration across accessors
Engine	Reusable logic with no storage concerns (TokenEngine signs JWTs)
Accessor	EF Core data access

Each layer sits behind an interface in DI.

Generated API client

The backend's OpenAPI document is fed to NSwag, which generates src/api/generated.ts: one typed client per controller, with types matching the C# DTOs.

Regenerate after changing a controller or DTO, with the backend running:

cd frontend && npm run generate-api

Getting started

Requires the .NET 10 SDK, Node 22+, and Docker.

1. Start SQL Server

docker run -e "ACCEPT_EULA=Y" -e "MSSQL_SA_PASSWORD=CatchIQ_Dev1!" \
  -p 1434:1433 --name catchiq-mssql \
  -d mcr.microsoft.com/mssql/server:2022-latest

Later runs: docker start catchiq-mssql.

2. Create backend/appsettings.Development.json (gitignored; appsettings.json holds placeholders):

{
  "ConnectionStrings": {
    "DefaultConnection": "Server=localhost,1434;Database=catchiq-mssql;User Id=sa;Password=CatchIQ_Dev1!;TrustServerCertificate=True;"
  },
  "Jwt": {
    "Secret": "replace-with-a-long-random-signing-key",
    "Issuer": "CatchIQ",
    "Audience": "CatchIQUsers"
  }
}

3. Set the Anthropic API key (optional — everything but Insights works without it):

cd backend && dotnet user-secrets set "Anthropic:ApiKey" "your-api-key"

4. Apply migrations:

cd backend && dotnet ef database update

5. Set up the frontend env (also gitignored; without it every API call hits the wrong URL):

cd frontend && cp .env.example .env

6. Run both:

cd backend && dotnet run --launch-profile http   # http://localhost:5045
cd frontend && npm install && npm run dev        # http://localhost:5173

CI

.github/workflows/ci.yml builds both halves on every push and PR to main: dotnet build -c Release against CatchIQ.slnx, and npm ci → lint → npm run build (which typechecks, unlike the dev server). Client generation is excluded, since NSwag needs a running backend.