[agent,nextjs,react] move Agent from constantinius, integrate deploy, delete old NextJS agent (#1165)

* [agent,nextjs,react] move Agent from constantinius, integrate deploy, delete old NextJS agent

* fixes

* use AGENT_URL_STAGING in staging and local

Author: maxkosty

.local.env.base

@@ -2,6 +2,7 @@ SENTRY_ORG= # ** Fill in **
 # -- line to prevent conflicts --
 SENTRY_AUTH_TOKEN=${__ENV__} # set in ~/.zshrc, must have permissions to your org
 GCP_PROJECT=sales-engineering-sf
+GCP_REGION=us-central1
 DB_CLOUD_SQL_CONNECTION_NAME=${__GCP_SECRET__}
 DB_DATABASE=hardwarestore
 DB_HOST=${__GCP_SECRET__}
@@ -14,7 +15,8 @@ SPRINGBOOT_URL=${__IF_DEPLOYING__(spring-boot, http://localhost:$SPRINGBOOT_LOCA
 ASPNETCORE_URL=${__IF_DEPLOYING__(aspnetcore, http://localhost:$ASPNETCORE_LOCAL_PORT, ${__GCP_SECRET__(ASPNETCORE_URL_STAGING)})}
 LARAVEL_URL=${__IF_DEPLOYING__(laravel, http://localhost:$LARAVEL_LOCAL_PORT, ${__GCP_SECRET__(LARAVEL_URL_STAGING)})}
 RUBYONRAILS_URL=${__IF_DEPLOYING__(ruby-on-rails, http://localhost:$RUBYONRAILS_LOCAL_PORT, ${__GCP_SECRET__(RUBYONRAILS_URL_STAGING)})}
-OPENAI_API_KEY=${__GCP_SECRET__}
+AGENT_URL=${__IF_DEPLOYING__(agent, http://localhost:$AGENT_LOCAL_PORT, ${__GCP_SECRET__(AGENT_URL_STAGING)})}
+MCP_URL=${__GCP_SECRET__} # Using production for now
 
 # Note prefix change in _.env.template
 REACT_DSN= # ** Fill in **
@@ -119,3 +121,9 @@ SPRINGBOOTOTLP_SENTRY_PROJECT=spring-boot-otlp
 # -- line to prevent conflicts --
 SPRINGBOOTOTLP_CLOUD_GCP_SQL_ENABLED=false
 SPRINGBOOTOTLP_RELEASE=empower.springboototlp@${__DYNAMIC_VERSION__}
+
+AGENT_DSN=${__GCP_SECRET__(AGENT_DSN_STAGING)}
+AGENT_LOCAL_PORT=8093
+# -- line to prevent conflicts --
+AGENT_OPENAI_API_KEY=${__GCP_SECRET__} # owned by EPD
+AGENT_SENTRY_ENVIRONMENT=local

README.md

@@ -186,10 +186,6 @@ gcloud app logs tail -s <SERVICE>
 
 `gcloud app deploy` does not support `--update-env-vars RELEASE=$RELEASE` like `gcloud run deploy` does with Cloud Run
 
-## AI Suggestions (nextjs -> flask)
-
-1. Run next and flask (./deploy --env=local next flask)
-2. Get suggestion button should show automatically
 
 ## Caches & Queues (flask)
 
@@ -226,4 +222,3 @@ When the script is terminated (e.g., with Ctrl+C), it performs cleanup to ensure
 - Once you have Vercel access run `vercel dev` command from the `empower` repo
 - Following the prompts, select the Sentry org from and the empower project. A vercel config file will be created the first time you do this.
 - App should run on `localhost:3000`
-

_bin/ensure_docker.sh

@@ -0,0 +1,58 @@
+#!/bin/bash
+
+# Ensure Docker (via Colima) is available for local development
+# Usage: source this script or call it directly
+#   source ensure_docker.sh  # exports DOCKER_HOST to current shell
+#   ./ensure_docker.sh       # just verifies docker is working
+
+set -e
+
+# Set Docker host to Colima's socket (must be set before any docker commands)
+export DOCKER_HOST="unix://${HOME}/.colima/default/docker.sock"
+
+ensure_docker() {
+  # Check for Homebrew
+  if ! command -v brew &> /dev/null; then
+    echo "Error: Homebrew is required to install dependencies. Install from https://brew.sh"
+    exit 1
+  fi
+
+  # Install Docker CLI if not present
+  if ! command -v docker &> /dev/null; then
+    echo "Installing Docker CLI..."
+    brew install docker
+  fi
+
+  # Install docker-compose if not present
+  if ! command -v docker-compose &> /dev/null; then
+    echo "Installing docker-compose..."
+    brew install docker-compose
+  fi
+
+  # Install Colima if not present
+  if ! command -v colima &> /dev/null; then
+    echo "Installing Colima..."
+    brew install colima
+  fi
+
+  # Start Colima if not running or socket doesn't exist
+  if ! colima status &> /dev/null || [ ! -S "${HOME}/.colima/default/docker.sock" ]; then
+    echo "Starting Colima..."
+    colima start
+  fi
+
+  # Verify Docker is working
+  echo "Verifying Docker connection (DOCKER_HOST=$DOCKER_HOST)..."
+  if ! docker info &> /dev/null; then
+    echo "Error: Docker not responding. Trying to restart Colima..."
+    colima stop 2>/dev/null || true
+    colima start
+    if ! docker info &> /dev/null; then
+      echo "Error: Docker still not responding after restarting Colima"
+      exit 1
+    fi
+  fi
+  echo "Docker is ready."
+}
+
+ensure_docker

agent/Dockerfile

@@ -0,0 +1,30 @@
+# Use Python 3.11 slim image
+FROM python:3.11-slim
+
+# Set working directory
+WORKDIR /app
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y \
+    gcc \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy requirements first for better caching
+COPY requirements.txt .
+
+# Install Python dependencies
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy application code
+COPY . .
+
+# Create non-root user
+RUN useradd --create-home --shell /bin/bash app \
+    && chown -R app:app /app
+USER app
+
+# Expose port (Cloud Run uses PORT env var, defaults to 8080)
+EXPOSE $PORT
+
+# Run the application
+CMD ["python", "main.py"]

agent/Makefile

@@ -0,0 +1,46 @@
+.PHONY: install format lint test clean dev-setup
+
+# Install dependencies
+install:
+	pip install -r requirements.txt
+
+# Development setup
+dev-setup: install
+	pre-commit install
+	@echo "Development environment setup complete!"
+	@echo "Auto import sorting and formatting will now run on git commits."
+
+# Format code
+format:
+	isort . --profile black --line-length 88
+	black . --line-length 88
+
+# Lint code
+lint:
+	flake8 . --max-line-length 88 --extend-ignore E203,W503
+	mypy . --ignore-missing-imports
+	isort . --profile black --line-length 88 --check-only
+
+# Run tests
+test:
+	pytest
+
+# Clean up generated files
+clean:
+	find . -type d -name "__pycache__" -delete
+	find . -type f -name "*.pyc" -delete
+	find . -type d -name "*.egg-info" -exec rm -rf {} +
+	find . -type d -name ".pytest_cache" -delete
+	find . -type d -name ".mypy_cache" -delete
+
+# Run the application
+run:
+	python main.py
+
+# Run with Docker
+docker-run:
+	docker-compose up --build
+
+# Sort imports manually
+sort-imports:
+	isort . --profile black --line-length 88

agent/README.md

@@ -0,0 +1,155 @@
+# Simple Plant Care AI Agent
+
+A simple FastAPI-based plant care AI assistant powered by the OpenAI Agents SDK. Just provide a plant name and get personalized care advice!
+
+## Features
+
+- 🌱 **Simple Plant Care Agent**: Get care advice for any plant by name
+- 🛠️ **One Tool**: Basic plant care database with common houseplants
+- 🚀 **FastAPI Backend**: Modern, fast API with automatic docs
+- 📚 **Interactive API Docs**: Built-in Swagger UI at `/docs`
+- 🎯 **Minimal Dependencies**: Bare bones implementation following OpenAI Agents SDK
+
+## Quick Start
+
+### Prerequisites
+
+- Python 3.11+
+- OpenAI API key
+
+### Installation
+
+1. **Clone the repository**
+
+   ```bash
+   git clone <repository-url>
+   cd empower-plant-agent-demo
+   ```
+
+2. **Set up environment variables**
+
+   ```bash
+   export OPENAI_API_KEY=sk-your-api-key-here
+   ```
+
+3. **Install dependencies**
+
+   ```bash
+   pip install -r requirements.txt
+   ```
+
+4. **Run the application**
+   ```bash
+   python main.py
+   ```
+
+The API will be available at `http://localhost:8000` with interactive docs at `http://localhost:8000/docs`.
+
+## Usage
+
+### API Endpoint
+
+**POST** `/api/v1/plant-care`
+
+Request body:
+
+```json
+{
+  "plant_name": "pothos"
+}
+```
+
+Response:
+
+```json
+{
+  "response": "🌱 **Pothos Care Guide**\n\n**Watering**: Water when top inch of soil is dry (every 1-2 weeks)\n**Light**: Bright, indirect light\n**Pro Tip**: Very forgiving plant, great for beginners. Can tolerate low light.\n\n**General Care Reminders**:\n• Check soil moisture before watering\n• Ensure good drainage\n• Watch for signs of overwatering (yellow leaves)\n• Rotate plant occasionally for even growth",
+  "agent_name": "PlantCareAgent"
+}
+```
+
+### Direct Python Usage
+
+```python
+import asyncio
+from app.agents.simple_plant_agent import process_plant_query
+
+async def main():
+    advice = await process_plant_query("snake plant")
+    print(advice)
+
+asyncio.run(main())
+```
+
+### Example with curl
+
+```bash
+curl -X POST "http://localhost:8000/api/v1/plant-care" \
+     -H "Content-Type: application/json" \
+     -d '{"plant_name": "monstera"}'
+```
+
+## Supported Plants
+
+The agent has specific care information for:
+
+- Pothos
+- Snake Plant
+- Monstera
+- Fiddle Leaf Fig
+- Peace Lily
+- Rubber Plant
+- Succulents
+
+For other plants, it provides general care advice and suggests trying the supported plants.
+
+## Project Structure
+
+```
+app/
+├── agents/
+│   └── simple_plant_agent.py    # Main agent with OpenAI Agents SDK
+├── tools/
+│   └── simple_plant_tool.py     # Plant care tool/function
+└── api/
+    ├── models.py                # Pydantic request/response models
+    └── routes.py         # FastAPI routes
+
+main.py                          # FastAPI application entry point
+simple_example.py               # Direct usage example
+config.py                       # Configuration management
+```
+
+## Configuration
+
+Set these environment variables:
+
+- `OPENAI_API_KEY`: Your OpenAI API key (required)
+- `light_model`: Model for the agent (default: "gpt-4o-mini")
+- `API_HOST`: API host (default: "0.0.0.0")
+- `PORT`: API port (default: 8000)
+
+## Development
+
+Run the simple example:
+
+```bash
+python simple_example.py
+```
+
+Check the API docs:
+
+```bash
+python main.py
+# Visit http://localhost:8000/docs
+```
+
+## OpenAI Agents SDK
+
+This project follows the [OpenAI Agents SDK](https://openai.github.io/openai-agents-python/) patterns:
+
+- **Agent**: Simple plant care assistant with instructions
+- **Tool**: Function tool for plant care advice
+- **Runner**: Handles the agent execution loop
+
+The implementation is minimal and focused, demonstrating the core concepts without complexity.

agent/app/__init__.py

@@ -0,0 +1 @@
+"""App package containing API and agents modules."""