🤖 aqua-os-crew

AI agent crews for water polo club operations — match prep, enrollment, progress reviews, and more.

Architecture

Python FastAPI service running CrewAI agents via OpenRouter/Gemini LLMs. Each crew handles a specific domain of club operations.

crews/
  match_prep/          → pre-game analysis & strategy
  enrollment/          → new player onboarding
  progress_review/     → player development tracking
  season_plan/         → season scheduling & goals
  injury_response/     → injury management workflows
tools/
  calendar_tool.py     → calendar integration tool
main.py                → FastAPI entrypoint

Prerequisites

• Python 3.11+
• OpenRouter API key or Gemini API key

Run Locally

pip install -r requirements.txt
uvicorn main:app --port 8001

Server starts on http://localhost:8001

Environment Variables

Variable	Description	Required
OPENROUTER_API_KEY	OpenRouter API key	One of these
GEMINI_API_KEY	Google Gemini API key	One of these

Docker

docker build -t aqua-os-crew .
docker run -p 8001:8001 -e OPENROUTER_API_KEY=sk-... aqua-os-crew

Related Repos

Repo	Description
aqua-os-backend	Go REST API
aqua-os-web	React frontend
aqua-os-calendar	Game calendar service
aqua-os-infrastructure	Terraform AWS infra

License

GPL-3.0

Metrics

Prometheus metrics are available via metrics.py. To enable, call in your FastAPI app:

from metrics import setup_metrics
setup_metrics(app)

Exposes auto-instrumented HTTP metrics at /metrics.