Sequential Questioning MCP Server

A specialized server that enables LLMs (Large Language Models) to gather specific information through sequential questioning. This project implements the MCP (Model Control Protocol) standard for seamless integration with LLM clients.

Project Status

🎉 Version 1.0.0 Released 🎉

The Sequential Questioning MCP Server is now complete and ready for production deployment. All planned features have been implemented, tested, and documented.

Features

Sequential Questioning Engine: Generates contextually appropriate follow-up questions based on previous responses
MCP Protocol Support: Full implementation of the MCP specification for integration with LLMs
Robust API: RESTful API with comprehensive validation and error handling
Vector Database Integration: Efficient storage and retrieval of question patterns
Comprehensive Monitoring: Performance metrics and observability with Prometheus and Grafana
Production-Ready Deployment: Kubernetes deployment configuration with multi-environment support
High Availability: Horizontal Pod Autoscaler and Pod Disruption Budget for production reliability
Security: Network policies to restrict traffic and secure the application

Documentation

Getting Started

Prerequisites

Python 3.10+
Docker and Docker Compose (for local development)
Kubernetes cluster (for production deployment)
PostgreSQL 15.4+
Access to a Qdrant instance

Quick Start

The easiest way to get started is to use our initialization script:

./scripts/initialize_app.sh

This script will:

Check if Docker is running
Start all necessary containers with Docker Compose
Run database migrations automatically
Provide information on how to access the application

The application will be available at http://localhost:8001

Local Development

Clone the repository

git clone https://github.com/your-organization/sequential-questioning.git
cd sequential-questioning

Install dependencies
```
pip install -e ".[dev]"
```

Set up environment variables

cp .env.example .env
# Edit .env file with your configuration

Run the development server
```
uvicorn app.main:app --reload
```

Docker Deployment

docker-compose up -d

Database Setup

If you're starting the application manually, don't forget to run the database migrations:

export DATABASE_URL="postgresql://postgres:postgres@localhost:5432/postgres"
bash scripts/run_migrations.sh

Kubernetes Deployment

Development Environment
```
kubectl apply -k k8s/overlays/dev
```
Staging Environment
```
kubectl apply -k k8s/overlays/staging
```
Production Environment
```
kubectl apply -k k8s/overlays/prod
```

See the Final Deployment Plan and Operational Runbook for detailed instructions.

Monitoring

Access Prometheus and Grafana dashboards for monitoring:

kubectl port-forward -n monitoring svc/prometheus 9090:9090
kubectl port-forward -n monitoring svc/grafana 3000:3000

CI/CD Pipeline

Automated CI/CD pipeline with GitHub Actions:

Continuous Integration: Linting, type checking, and testing
Continuous Deployment: Automated deployments to dev, staging, and production
Deployment Verification: Automated checks post-deployment

Testing

Run the test suite:

pytest

Run performance tests:

python -m tests.performance.test_sequential_questioning_load

Troubleshooting

Database Tables Not Created

If the application is running but the database tables don't exist:

Make sure the database container is running

Run the database migrations manually:

export DATABASE_URL="postgresql://postgres:postgres@localhost:5432/postgres"
bash scripts/run_migrations.sh

Pydantic Version Compatibility

If you encounter the error pydantic.errors.PydanticImportError: BaseSettings has been moved to the pydantic-settings package, ensure that:

The pydantic-settings package is included in your dependencies
You're importing BaseSettings from pydantic_settings instead of directly from pydantic

This project uses Pydantic v2.x which moved BaseSettings to a separate package.

Contributing

Contributions are welcome! Please see CONTRIBUTING.md for guidelines.

License

MIT License

Contact

For support or inquiries, contact support@example.com