Install / Deploy
Enterprise AI Agent Deployment
Prerequisites
| Requirement | Details |
|---|---|
| Python | 3.8 or higher |
| OS | Windows or Linux |
| RAM | Minimum 8 GB (16 GB recommended for production) |
| SAP AI Core | Access credentials required |
| HANA Database | Connection per tenant |
| SharePoint | Integration credentials (optional) |
| Microsoft Graph API | Required for calendar features |
Step 1 — Set Up Python Environment
python -m venv venv
# Windows
.\venv\Scripts\Activate.ps1
# Linux
source venv/bin/activate
pip install -r requirements.txt
Step 2 — Initialize Database
python -c "from database import init_all_engines; init_all_engines()"
Step 3 — Build & Push Docker Image to Azure Container Registry
# Login to Azure
az login
az acr login -n wblnd
# Build and push image
docker build -t wblnd.azurecr.io/agent-cherry-backend:v3.0.1 .
docker push wblnd.azurecr.io/agent-cherry-backend:v3.0.1
Step 4 — Deploy to Cloud Foundry (SAP BTP)
# Login to Cloud Foundry (SSO)
cf8 login --sso
# If API endpoint is not shown, set it manually
cf8 api https://api.cf.eu10-004.hana.ondemand.com
# A URL will be displayed — paste it in your browser to get the SSO passcode
# (Use default origin or token-based login as applicable)
# Set Docker registry password
set CF_DOCKER_PASSWORD=your_password
# Push the application
cf8 push
Step 5 — Verify Deployment
| Check | URL |
|---|---|
| API Documentation | http://localhost:8000/docs |
| Health Endpoint | http://localhost:8000/health |
| Metrics Endpoint | http://localhost:8000/metrics |
Alternative — Run Locally Without Docker
# Development
uvicorn main:app --host 0.0.0.0 --port 8000 --reload
# Production (direct)
gunicorn main:app -w 4 -k uvicorn.workers.UvicornWorker --bind 0.0.0.0:8000
MCP Server Install/Deployment
Prerequisites
| Requirement | Who Provides It |
|---|---|
| Python 3.8+ | Developer workstation / CI environment |
requirements.txt | Available in the project repository |
| Environment variables (API keys, DB connection, FAISS path) | DevOps / Platform team |
FAISS index file (.faiss) | Data / ML team |
| Backend service URLs (dev / qa / prod) | Integration / Infrastructure team |
Step 1 — Create & Activate Virtual Environment
python -m venv venv
# Windows
.\venv\Scripts\Activate.ps1
# macOS / Linux
source venv/bin/activate
Step 2 — Install Dependencies
pip install -r requirements.txt
Step 3 — Set Minimum Required Environment Variables
FAISS_INDEX_PATH=faiss_index/soa_dev/index.faiss
DB_CONN=<your-database-connection-string>
See Section 2 (How to Configure) for the full list of supported variables.
Step 4 — Start the Server
python main.py
Key Files to Know
| File | Purpose |
|---|---|
main.py | Application entry point |
requirements.txt | Python dependency list |
manifest.yaml | Tool and service metadata |
db_config.py | Database configuration |
Verify the Server is Running
| Check | Endpoint |
|---|---|
| Health / Status | GET /status |
| Tool Discovery | POST /mcp/{env} with tools/list method |