Promptstudio

Created By

BodieCoding7 months ago

PromptStudio - An MCP server and Application for managing AI prompts with collections, variables, execution history, and import/export capabilities.

# ai

# mcp

Overview Content Tools Comments

Content

Professional AI Prompt Management - Organize, test, and manage your AI prompts with ease.

Prompt Studio is a clean, modern web application that helps you organize your AI prompts into collections, define variables for dynamic content, and test prompts with different values - just like professional API testing tools do for web services.

✨ Features

📁 Collections: Organize prompts into logical groups
⚡ Dynamic Variables: Use {{variable}} syntax for reusable prompts
🧪 Test & Execute: Run prompts with different variable values
📊 Execution History: Track when and how prompts were used
🎨 Modern UI: Clean, professional interface
🔗 Variable Auto-Detection: Automatically extracts variables from prompt content
📋 Variable Collections: Batch execute prompts with CSV data sets - like data-driven testing in professional API tools
📥 CSV Templates: Download properly formatted CSV templates for your prompt variables
🚀 Batch Processing: Execute prompts against multiple variable sets simultaneously
📤 Results Export: Export comprehensive batch execution results to CSV
🤖 MCP Integration: Model Context Protocol server for AI agent integration (GitHub Copilot, Claude Desktop, etc.)
🔌 API Access: RESTful API endpoints for programmatic access to all functionality

🏗️ Architecture

Built with Clean Architecture principles:

PromptStudio/
├── Domain/              # Core business entities
│   ├── Collection.cs
│   ├── PromptTemplate.cs
│   ├── PromptVariable.cs
│   ├── PromptExecution.cs
│   └── VariableCollection.cs
├── Services/            # Business logic
│   └── PromptService.cs
├── Data/                # Entity Framework
│   └── PromptStudioDbContext.cs
└── Pages/               # Razor Pages UI
    ├── Collections/
    ├── Prompts/
    └── VariableCollections/

🚀 Quick Start

Prerequisites

.NET 8 SDK
SQL Server (Docker or local)

Setup with Docker SQL Server

Start SQL Server in Docker:

docker pull mcr.microsoft.com/mssql/server:2019-latest
docker run -e 'ACCEPT_EULA=Y' -e 'SA_PASSWORD=YourPassword123!' -p 1433:1433 --name sqlserver -d mcr.microsoft.com/mssql/server:2019-latest

Clone and Run:

git clone <your-repo>
cd PromptStudio
dotnet run

Open Browser: Navigate to http://localhost:5131

🤖 MCP Server Setup (GitHub Copilot Integration)

PromptStudio includes a Model Context Protocol (MCP) server that enables AI agents like GitHub Copilot to access and manage your prompts programmatically.

Prerequisites for MCP Server

Node.js 18+ (for MCP server)
PromptStudio application running (see setup above)

Install and Run MCP Server

Navigate to MCP server directory:
```
cd mcp-server
```
Install dependencies:
```
npm install
```
Build the TypeScript code:
```
npm run build
```
Start the MCP server:
```
npm start
```
Or for development with auto-reload:
```
npm run dev
```

Configure with GitHub Copilot

To use the MCP server with GitHub Copilot, add this configuration to your VS Code settings:

Open VS Code Settings (JSON)
Add the MCP server configuration:

{
  "github.copilot.chat.experimental.codeGeneration.instructions": [
    {
      "text": "You have access to PromptStudio via MCP server for professional prompt management."
    }
  ],
  "mcp.servers": {
    "promptstudio": {
      "command": "node",
      "args": ["path/to/your/mcp-server/dist/index.js"],
      "env": {
        "PROMPTSTUDIO_BASE_URL": "http://localhost:5131"
      }
    }
  }
}

Configure with Claude Desktop

Add to your Claude Desktop MCP configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "promptstudio": {
      "command": "node",
      "args": ["path/to/your/mcp-server/dist/index.js"],
      "env": {
        "PROMPTSTUDIO_BASE_URL": "http://localhost:5131"
      }
    }
  }
}

MCP Tools Available

The MCP server provides these tools for AI agents:

csv_template_generate - Generate CSV templates for prompt variables
variable_collection_create_from_csv - Create variable collections from CSV data
variable_collections_list - List variable collections for prompts
prompt_execute - Execute prompts with variables
batch_execute - Batch execute prompts with variable collections

Example MCP Usage

Once configured, you can ask GitHub Copilot or Claude:

"Generate a CSV template for prompt ID 1 in PromptStudio"

"Create a variable collection from this CSV data for my code review prompt"

"List all variable collections for prompt template 2"

The AI agent will use the MCP server to interact with your PromptStudio instance automatically.

Troubleshooting MCP Setup

Ensure PromptStudio is running on http://localhost:5131
Check Node.js version: node --version (must be 18+)
Verify MCP server build: Check dist/index.js exists after npm run build
Test API endpoints: Visit http://localhost:5131/api/mcp/prompts in browser
Check logs: MCP server logs will show connection status and errors

First Steps

Create a Collection - Like a project folder for related prompts
Add Prompts - Write prompts with {{variables}} for dynamic content
Execute & Test - Run prompts with different variable values
Batch Processing - Create Variable Collections for testing multiple scenarios
Download Templates - Get CSV templates with your prompt variables
Upload Data Sets - Fill CSV with multiple test scenarios and upload
Batch Execute - Run prompts against all variable sets simultaneously
Export Results - Download comprehensive results for analysis

📝 Example Prompts

Code Review

Please review the following {{language}} code for {{focus_area}}:

```{{language}}
{{code}}

Focus on:

Code quality
Performance
Security
Best practices


### Content Writing

Write a {{tone}} {{content_type}} about {{topic}} for {{audience}}.

Requirements:

Length: {{length}}
Include: {{key_points}}
Avoid: {{avoid_topics}}


### Batch Testing with Variable Collections

Create a CSV file to test multiple scenarios at once:

**CSV Template** (auto-generated from your prompt):
```csv
language,code
javascript,"function greet(name) { return 'Hello ' + name; }"
python,"def greet(name): return f'Hello {name}'"
java,"public class Greeting { public static String greet(String name) { return ""Hello "" + name; } }"
typescript,"function greet(name: string): string { return `Hello ${name}`; }"

Workflow:

Download CSV Template → Get properly formatted headers
Fill with Data → Add multiple test scenarios in Excel/editor
Upload & Execute → Batch process all variable sets
Export Results → Download comprehensive results with resolved prompts

📸 Screenshots

Main Dashboard

PromptStudio Dashboard Clean, modern dashboard showing collections overview with prompt counts and recent activity

Creating New Prompts

New Prompt Creation Intuitive prompt creation interface with variable auto-detection and syntax highlighting

Prompt Execution

Prompt Execution Execute prompts with variable substitution and see resolved output in real-time

Variable Collections Management

Variable Collections Manage multiple variable sets for batch testing and data-driven prompt execution

Batch Variable Collection

Advanced batch processing interface for executing prompts against multiple variable collections simultaneously

CSV Import for Variable Collections

CSV Import Import CSV data to create variable collections - Excel/CSV workflow integration for enterprise-scale testing

MCP Integration with AI Assistants

MCP Tools Model Context Protocol integration enables AI assistants like GitHub Copilot and Claude Desktop to access PromptStudio functionality through natural language commands

🔥 Variable Collections - Batch Processing Power

Transform single prompt testing into enterprise-scale batch processing!

Key Benefits

10x Faster Testing - Process hundreds of scenarios vs. manual one-by-one execution
Systematic Quality Assurance - Ensure prompts work across diverse scenarios
Data-Driven Insights - Export results for analysis and optimization
Excel/CSV Integration - Use familiar tools to manage test data

How It Works

Create a Prompt with variables like {{language}} and {{code}}
Download CSV Template - Auto-generated with correct headers:
```
language,code
```

Fill with Test Data in Excel or any CSV editor:

language,code
javascript,"function hello() { console.log('Hello World'); }"
python,"def hello(): print('Hello World')"
java,"public class Hello { public static void main(String[] args) { System.out.println(""Hello World""); } }"

Upload & Execute - Process all rows automatically
Export Results - Get comprehensive CSV with:
- Original variables for each test case
- Resolved prompts with variables substituted
- Execution metadata and timestamps
- Success/failure status for each scenario

Perfect For

A/B Testing - Compare prompt variations across multiple inputs
Quality Assurance - Test prompts against edge cases and diverse scenarios
Data Processing - Batch process large datasets through AI workflows
Performance Testing - Validate prompt performance at scale

🎯 Use Cases

Development Teams: Standardize code review prompts, batch test across multiple codebases
Content Creators: Template-driven content generation with A/B testing scenarios
Support Teams: Consistent customer response templates with variable customer data
Researchers: Structured data analysis prompts with batch data processing
Educators: Reusable teaching and assessment prompts with student cohort testing
QA Teams: Systematic prompt testing across diverse scenarios and edge cases
Data Scientists: Batch process datasets through AI models with variable parameters

🛠️ Technology Stack

Backend: ASP.NET Core 8, Entity Framework Core
Frontend: Razor Pages, Bootstrap 5, Bootstrap Icons
Database: SQL Server
Architecture: Clean Architecture, Repository Pattern

🔧 Configuration

Update appsettings.json with your database connection:

{
  "ConnectionStrings": {
    "DefaultConnection": "Server=localhost,1433;Database=PromptStudio;User Id=sa;Password=YourPassword123!;TrustServerCertificate=true;"
  }
}

📊 Database Schema

Collections (id, name, description, created_at, updated_at)
├── PromptTemplates (id, collection_id, name, description, content, created_at, updated_at)
    ├── PromptVariables (id, template_id, name, description, default_value, type, created_at)
    ├── VariableCollections (id, template_id, name, description, variable_sets_json, created_at, updated_at)
    └── PromptExecutions (id, template_id, variable_collection_id, resolved_prompt, variable_values, executed_at, ...)