Phase 2 + 3 from Google Jules

backend/.env.example

@@ -0,0 +1,40 @@
+# Environment variables for the Backend Service
+
+# -- Database Configuration (Supabase PostgreSQL) --
+# Format: postgresql://postgres:[YOUR-PASSWORD]@[AWS-REGION].pooler.supabase.com:6543/postgres
+SUPABASE_DATABASE_URL="postgresql://postgres:[email protected]:6543/postgres"
+
+# -- OpenAI Configuration --
+OPENAI_API_KEY="sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
+
+# -- Service Security Note --
+# Client authentication is handled by an upstream Supabase Edge Function (`auth-gateway`).
+# Refer to `backend/docs/supabase_edge_functions.md` for details.
+
+# -- Tool Credentials (Example) --
+MOCK_WEATHER_TOOL_API_KEY="test_weather_key" 
+
+# -- Logging Configuration --
+# LOG_LEVEL="INFO" # Optional: DEBUG, INFO, WARNING, ERROR, CRITICAL. Defaults to INFO.
+
+# -- OpenTelemetry Configuration --
+# Endpoint for the OTLP collector (e.g., Jaeger, Grafana Tempo, OpenTelemetry Collector).
+# For HTTP exporter, typically ends with /v1/traces.
+OTEL_EXPORTER_OTLP_ENDPOINT="http://localhost:4318/v1/traces"
+# Service name to identify this backend in traces.
+OTEL_SERVICE_NAME="ai_assistant_backend"
+# Optional: Set to "true" to print traces to console (for debugging, very verbose)
+# OTEL_EXPORTER_OTLP_TRACES_EXPORTER="console" 
+
+# -- Optional: Agent Behavior --
+# DEFAULT_AGENT_PERSONA_NAME="greeter"
+
+# -- Optional: Uvicorn/FastAPI settings --
+# HOST="0.0.0.0"
+# PORT="8000"
+# WORKERS="4"
+
+# Note: 
+# - Rename this file to .env and fill in your actual values.
+# - Add .env to your .gitignore file.
+# - For testing, consider a separate .env.test file.

backend/.env.local.example

@@ -0,0 +1,47 @@
+# Local Docker Development Environment Variables for the Backend Service
+# Rename this to .env.local (or a name referenced in docker-compose.yml)
+
+# --- Database Configuration ---
+# For local development with Docker Compose, you might connect to:
+# 1. A cloud-hosted Supabase instance (recommended for ease).
+# 2. A local Supabase instance running via its own Docker setup (more complex, see Supabase docs).
+# 3. A standard PostgreSQL container (if not using Supabase-specific features heavily during dev).
+
+# Option 1: Connect to your cloud Supabase instance (most common for dev)
+# Replace with your actual Supabase project's connection string.
+SUPABASE_DATABASE_URL="postgresql://postgres:[email protected]:6543/postgres"
+
+# Option 2: Example if Supabase was running locally via Docker on default port 5432
+# This assumes the backend service can reach the Supabase stack on 'host.docker.internal' (from Docker Desktop)
+# or a specific service name if part of the same docker-compose network.
+# SUPABASE_DATABASE_URL="postgresql://postgres:[email protected]:5432/postgres"
+# Or if 'supabase_db' was a service name in docker-compose:
+# SUPABASE_DATABASE_URL="postgresql://postgres:your-local-supabase-password@supabase_db:5432/postgres"
+
+
+# --- OpenAI Configuration ---
+# Your API key for OpenAI services.
+# It's recommended to use a development-specific key if possible.
+OPENAI_API_KEY="sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
+
+
+# --- Tool Credentials (Example) ---
+# API key for the mock weather tool.
+MOCK_WEATHER_TOOL_API_KEY="test_weather_key"
+
+
+# --- Logging Configuration ---
+# Optional: Set the log level for the application (DEBUG, INFO, WARNING, ERROR, CRITICAL).
+# Defaults to INFO if not set in the application.
+LOG_LEVEL="DEBUG" # Often useful for local development
+
+
+# --- Mock A2A Server URL (if needed by backend to know its own mock's URL, usually not) ---
+# Generally, the A2A service discovers other agents. If the backend needs to point to the
+# mock A2A server for some reason (e.g., if an agent definition card in the backend
+# explicitly needs to point to the mock server by its Docker service name), you might add:
+# MOCK_A2A_SERVER_BASE_URL="http://mock_a2a_server:8001"
+# However, MOCK_AGENT_CARDS in a2a_communication_service.py already has http://localhost:8001,
+# which works when all services are mapped to localhost ports. If running purely container-to-container,
+# then service names would be used.
+```

backend/.env.test.example

@@ -0,0 +1,34 @@
+# Example .env.test file for configuring tests
+# Rename this to .env.test and fill in actual values if needed for your test environment.
+# python-dotenv (if used in your test setup) can load this file.
+
+# --- OpenAI API Key ---
+# For running integration tests that might make real calls to OpenAI (if not mocked).
+# Most unit/integration tests in this project mock the OpenAI client.
+# OPENAI_API_KEY="sk-your_actual_openai_api_key_for_testing"
+
+# --- Supabase Database URL for Integration Testing ---
+# Optional: If you want to run integration tests (`test_conversation_flow.py`, `test_a2a_flow.py`)
+# against a real Supabase PostgreSQL instance instead of the default SQLite.
+# The tests are configured to use this environment variable if set.
+# Format: postgresql://postgres:[YOUR-PASSWORD]@[AWS-REGION].pooler.supabase.com:6543/postgres
+# TEST_SUPABASE_DATABASE_URL="postgresql://postgres:[email protected]:6543/postgres"
+
+# --- Logging Level for Tests ---
+# Optional: Set the log level for test runs (e.g., DEBUG, INFO, WARNING).
+# The application's logging_config.py will use this if set.
+# LOG_LEVEL="DEBUG"
+
+# --- Authentication Note for Backend Tests ---
+# The backend's AG-UI endpoint now expects an "X-Auth-Validation-Status: success" header,
+# which is assumed to be set by an upstream Supabase Edge Function that validates the original
+# client's X-API-KEY.
+# Therefore, `EXPECTED_API_KEY` is NO LONGER configured directly in the backend's .env file
+# for runtime. For tests that hit these protected endpoints (like integration tests),
+# the test client directly sets the "X-Auth-Validation-Status: success" header.
+# The Supabase Edge Function itself would need an `EXPECTED_API_KEY_SECRET` configured in its own environment.
+
+# --- Other Test-Specific Settings ---
+# If your agent personas or other services load config from env vars for testing:
+# AGENT_PERSONA_DEFAULT_NAME="testGreeter"
+# MOCK_A2A_SERVER_URL="http://localhost:8001" # If the mock A2A server URL needs to be configurable for tests

backend/Dockerfile

@@ -0,0 +1,42 @@
+# Use an official Python runtime as a parent image
+FROM python:3.11-slim
+
+# Set environment variables for Python
+ENV PYTHONDONTWRITEBYTECODE 1
+ENV PYTHONUNBUFFERED 1
+
+# Set the working directory in the container
+WORKDIR /app
+
+# Install system dependencies that might be needed by Python packages
+# Example: psycopg2 might need libpq-dev if not using the binary version
+# RUN apt-get update && apt-get install -y --no-install-recommends \
+#     build-essential \
+#     libpq-dev \
+#  && rm -rf /var/lib/apt/lists/*
+
+# Copy the requirements file into the container at /app
+COPY requirements.txt .
+
+# Install Python dependencies
+# --no-cache-dir: Disables the pip cache, reducing image size.
+# --upgrade pip: Ensures pip is up-to-date.
+RUN pip install --no-cache-dir --upgrade pip -r requirements.txt
+
+# Copy the rest of the backend application code into the container at /app
+# This includes all subdirectories like services, routers, models, etc.
+COPY . .
+
+# Expose the port the app runs on
+# This should match the port Uvicorn is configured to use.
+EXPOSE 8000
+
+# Define the command to run the application
+# Use 0.0.0.0 to make the server accessible externally (from outside the container).
+# The --reload flag is typically for development; for production, you might remove it
+# or control it via an environment variable (e.g., APP_ENV=development).
+# For this Dockerfile, we'll include --reload for easier dev with docker-compose volume mounts.
+# If you want a production-ready image, create a separate Dockerfile or use multi-stage builds
+# to remove dev dependencies and disable reload.
+CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000", "--reload"]
+```