Merge branch 'feat/add-helm-charts' into 'main'

The Dockerfile sets these defaults:

```

ENV APP_HOST=0.0.0.0

ENV APP_PORT=9013

ENV MCP_SERVER_URL=http://127.0.0.1:8004/mcp

...

This deployment relies on a `.env` file that is loaded at runtime by Docker Compose.

```

GROQ_API_KEY=your-secret-key

OpenOP_BACKEND_OAI=http://example-backend

GROQ_API_KEY=your-groq-api-key

OEG_SERVICE_URL=http://your-oeg-api-url/oeg/1.0.0

GROQ_MODEL_NAME=qwen/qwen3-32b   

```

## Run

Start services

```

docker compose --env-file .env up -d

```

## Additional Info

All service ports are explicitly forwarded to the host to simplify local development, testing and debugging.

This allows you to interact with APIs and MCP Servers individually, without attaching to containers or using internal Docker networking tools.

This allows you to interact with APIs and MCP servers individually, without attaching to containers or using internal Docker networking tools.

| Service   | Host Port |

|-----------|-----------|

| mcp       | 8004      |

| ai-agent  | 9013      |

---

 No newline at end of file

version: "3.9"

services:

  api:

    image: labs.etsi.org:5050/oop/code/ai2/dummy_backend

    container_name: api

    environment:

      API_HOST: 0.0.0.0

      API_PORT: 8081

    ports:

      - "8081:8081"

    networks:

      - ai_net

  mcp:

    image: labs.etsi.org:5050/oop/code/ai2/mcp_module

    container_name: mcp

    environment:

      MCP_HOST: 0.0.0.0

      MCP_PORT: 8004

      OpenOP_BACKEND_OAI: ${OpenOP_BACKEND_OAI}

      OEG_SERVICE_URL: ${OEG_SERVICE_URL}

    ports:

      - "8004:8004"

    networks:

      - ai_net

    healthcheck:

      test: ["CMD", "python3", "-c", "import urllib.request; urllib.request.urlopen('http://localhost:8004/health')"]

      interval: 10s

      timeout: 5s

      retries: 5

      start_period: 10s

  ai-agent:

    image: labs.etsi.org:5050/oop/code/ai2/ai_agent

    container_name: ai-agent

    depends_on:

      - mcp

      mcp:

        condition: service_healthy

    environment:

      APP_HOST: 0.0.0.0

      APP_PORT: 9013

      MCP_SERVER_URL: http://mcp:8004/mcp

      GROQ_API_KEY: ${GROQ_API_KEY}

      GROQ_MODEL_NAME: openai/gpt-oss-20b

      GROQ_MODEL_NAME: ${GROQ_MODEL_NAME:-qwen/qwen3-32b}

    ports:

      - "9013:9013"

    networks:

# AI2 Helm Chart

Deploys the AI2 stack — MCP server and AI Agent — into a Kubernetes cluster.

## Components

| Component  | Description                                              | Default Port |

|------------|----------------------------------------------------------|-------------|

| mcp        | MCP server exposing CAMARA-compliant tools via FastMCP   | 8004        |

| ai-agent   | Quart backend accepting natural language prompts via Groq | 9013        |

The ai-agent waits for the MCP server to be healthy before starting (via initContainer).

## Prerequisites

- Kubernetes cluster

- Helm 3+

- Groq API key

- OEG service reachable from within the cluster

## Install

```bash

helm install ai2 ./helm/ai2 \

  --namespace oop \

  --create-namespace \

  --set secrets.groqApiKey=<your-groq-api-key> \

  --set config.mcp.oegServiceUrl=http://<oeg-host>/oeg/1.0.0

```

## Upgrade

```bash

helm upgrade ai2 ./helm/ai2 \

  --namespace oop \

  --set secrets.groqApiKey=<your-groq-api-key> \

  --set config.mcp.oegServiceUrl=http://<new-oeg-host>/oeg/1.0.0

```

## Uninstall

```bash

helm uninstall ai2 --namespace oop

```

## Key Values

| Value                           | Description                      | Default                  |

|---------------------------------|----------------------------------|--------------------------|

| `secrets.groqApiKey`            | Groq API key (required)          | `""`                     |

| `config.mcp.oegServiceUrl`      | Base URL of the OEG service      | `http://oeg/oeg/1.0.0`   |

| `config.aiAgent.groqModelName`  | Groq model identifier            | `qwen/qwen3-32b`         |

| `service.mcp.type`              | Kubernetes service type for MCP  | `NodePort`               |

| `service.mcp.nodePort`          | NodePort for MCP                 | `32004`                  |

| `service.aiAgent.nodePort`      | NodePort for ai-agent            | `32013`                  |

| `images.mcp.tag`                | MCP image tag                    | `latest`                 |

| `images.aiAgent.tag`            | AI Agent image tag               | `latest`                 |

## Local Development (kind)

Build and load images locally instead of pulling from the registry:

```bash

docker build -t <your_mcp_dev_image>:latest ../../mcp_module

docker build -t <your_ai_agent_dev_image>:latest ../../ai_agent

kind load docker-image <your_mcp_dev_image>:latest --name <your-cluster>

kind load docker-image <your_ai_agent_dev_image>:latest --name <your-cluster>

```

Then install with `pullPolicy: IfNotPresent` to use the local images:

```bash

helm install ai2 ./helm/ai2 \

  --namespace oop \

  --create-namespace \

  --set secrets.groqApiKey=<your-groq-api-key> \

  --set config.mcp.oegServiceUrl=http://<oeg-host>/oeg/1.0.0 \

  --set images.mcp.pullPolicy=IfNotPresent \

  --set images.aiAgent.pullPolicy=IfNotPresent

```

## Test the Deployment

```bash

# Health checks

curl http://<node-ip>:32004/health   # MCP

curl http://<node-ip>:32013/health   # AI Agent

# Send a natural language prompt

curl -X POST http://<node-ip>:32013/groq-mcp \

  -H "Content-Type: application/json" \

  -d '{"query": "Create a QoD session for device with public IP 12.1.0.20, application server IP 198.51.100.10, QoS profile qos-e, duration 3600 seconds"}'

```

apiVersion: v2

name: ai2

type: application

version: 1.0.0

appVersion: "1.0.0"

keywords:

  - ai2

  - oop

  - etsi

  - mcp

  - ai-agent

home: https://labs.etsi.org/rep/oop/code/ai2

maintainers:

  - name: OOP Team