About PromptForge
PromptForge is an MCP server in the RAG category: reusable generative prompt packages for Claude Code — scaffold entire project architectures on demand with /forge run.
The repository has 1 stars and 0 forks.
Install
git clone https://github.com/Stevekaplanai/promptforge-mcp-serverThis server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.
Ask AI
Ask AI about PromptForge
Powered by Claude · Grounded in docs
Security checks
- ✓README presentThe repository ships documentation.
5 checks have not been run yet.
README
PromptForge MCP Server
Intelligent prompt optimization middleware that enhances user prompts before execution.
Features
- 🎯 Smart Pattern Matching - Automatically detects prompt intent and applies relevant optimizations
- 📊 Analytics Tracking - Monitor optimization performance and patterns
- 🔧 Customizable Patterns - Add and modify optimization patterns for your specific needs
- 🚀 Domain-Specific Enhancements - Pre-configured patterns for marketing, data analysis, tax/accounting, and code generation
- 💾 External Service Integration - Uses JSONBin for pattern storage and Supabase for analytics
Setup
1. Install Dependencies
npm install
2. Configure External Services
JSONBin.io (Pattern Storage)
- Create account at https://jsonbin.io
- Create a new bin with initial patterns
- Copy your bin ID and API key
Supabase (Analytics)
- Create account at https://supabase.com
- Create a new project
- Create a table called
prompt_analyticswith this schema:- id (uuid, primary key)
- original_prompt (text)
- optimized_prompt (text)
- domain (text)
- confidence (float)
- modifications (json)
- user_feedback (json, nullable)
- performance_metrics (json, nullable)
- created_at (timestamp with timezone)
3. Environment Variables
Copy .env.example to .env and fill in your API credentials:
cp .env.example .env
4. Test Your Setup
npm test
Available Tools
optimize_prompt
Optimizes a prompt using intelligent pattern matching.
{
"tool": "optimize_prompt",
"arguments": {
"prompt": "Write an email about our services",
"domain": "marketing_copy",
"context": {
"company": "Your Company",
"audience": "B2B customers"
}
}
}
get_patterns
Retrieve current optimization patterns.
{
"tool": "get_patterns",
"arguments": {
"domain": "marketing_copy"
}
}
update_pattern
Update or add a pattern in the library.
{
"tool": "update_pattern",
"arguments": {
"domain": "custom_domain",
"pattern": {
"triggerKeywords": ["keyword1", "keyword2"],
"enhancements": [
{
"type": "role_addition",
"value": "You are an expert..."
}
]
}
}
}
get_analytics_summary
Get analytics summary for optimizations.
{
"tool": "get_analytics_summary",
"arguments": {
"timeRange": "week",
"domain": "marketing_copy"
}
}
test_optimization
Test optimization with before/after comparison.
{
"tool": "test_optimization",
"arguments": {
"prompt": "analyze our sales data",
"showDiff": true
}
}
Deployment to mcpify.ai
-
Create New MCP Server on mcpify.ai
- Name:
promptforge - Description: "Intelligent prompt optimization for consistent AI outputs"
- Name:
-
Upload Files
- Copy the contents of
server.jsto mcpify.ai - Add environment variables from your
.envfile
- Copy the contents of
-
Configure Tools
- The server automatically exposes all tools defined in
getTools()
- The server automatically exposes all tools defined in
Default Patterns
The server includes pre-configured patterns for:
- Marketing Copy - Copywriting, emails, landing pages, ads
- Data Analysis - Analytics, reports, insights, metrics
- Tax & Accounting - Financial planning, tax optimization, compliance
- Code Generation - Programming, debugging, implementation
- General - Fallback for unmatched prompts
For Schapira CPA Integration
This server is optimized for use with Schapira CPA's services:
{
"context": {
"company": "Schapira CPA",
"industry": "Tax and Accounting Services",
"audience": "Mid-market owner-operated businesses",
"tone": "professional"
}
}
License
MIT - Feel free to customize for your needs!
README copy captured when this server was indexed; it may lag the repository. The original is authoritative.