Promptstudio

What is PromptStudio?

PromptStudio is an AI-powered prompt management and workflow orchestration platform built on modern .NET architecture. It provides tools for creating, managing, and executing advanced AI agent workflows with specialized templates for enterprise automation, code analysis, debugging, content generation, and quality assurance.

How to use PromptStudio?

Clone the repository, then use Docker Compose (docker-compose up -d) for production or local development with dotnet build and dotnet run. Start the optional MCP server via cd mcp-server && npm install && npm start. Access the web UI at http://localhost:5131 (local) or http://localhost:5000 (Docker).

Key features of PromptStudio

• Modern .NET architecture with clean layered separation
• AI Agent Workflows for complex multi-step automation
• Enterprise-grade code analysis and refactoring
• Intelligent debugging and problem-solving frameworks
• Multi-modal content generation for multi-platform campaigns
• Workflow orchestration with error handling and monitoring

Use cases of PromptStudio

• Automate code review and refactoring with business context
• Diagnose and resolve technical issues systematically
• Generate multi-platform content with brand consistency
• Orchestrate complex AI workflows with fallback strategies
• Generate tests and integrate QA automation into CI/CD

FAQ from PromptStudio

What are the prerequisites to run PromptStudio?

You need .NET 8.0 SDK, SQL Server (LocalDB or full), Node.js 18+ (for MCP server), and optionally Docker & Docker Compose for containerized deployment.

How do I start the MCP server?

Navigate to the mcp-server directory, run npm install && npm start. The MCP server runs on port 3001.

Does PromptStudio support Docker deployment?

Yes, Docker Compose orchestrates the web application, SQL Server, and MCP server. Use docker-compose up -d for production or docker-compose -f docker-compose.yml -f docker-compose.dev.yml up for development with hot reload.

How can I execute prompts via the API?

Send a POST request to /api/prompts/execute with a JSON body containing id and variables. Example: curl -X POST http://localhost:5131/api/prompts/execute -H "Content-Type: application/json" -d '{"id": 1001, "variables": "{\"code_type\":\"microservice\"}"}'.