AI Prompt Management and Task-Master Integration System
Intelligent prompt management with context-aware suggestions, multi-project support, and seamless browser integration. Designed for developers, content creators, and teams who work with AI assistants across multiple projects.
- ๐ Template System - Variables, categories, and reusable prompts
- ๐ Secure Variables - AES-256 encrypted storage for API keys and sensitive data
- ๐งน Log Sanitization - Intelligently filter repetitive patterns from verbose logs
- ๐ฏ Context-Aware Suggestions - Smart prompt recommendations based on current task
- ๐ Usage Analytics - Track prompt effectiveness and success rates
- ๐ Search & Filter - Find prompts by category, tags, or content
- ๐ Multi-Project Support - Manage multiple Task-Master projects
- ๐ค Auto-Context Extraction - Automatically fills PRD locations, requirements, success criteria
- ๐ Project Registry - Manual and automatic project discovery
- โก One-Command Context -
pmcliinstantly copies perfect project context
- ๐ Cross-Platform - Works on Claude.ai, ChatGPT, GitHub, Notion, Linear, Slack
- ๐ซ Grammarly-Style Integration - Unobtrusive suggestions when you need them
- ๐ฑ Visual Project Picker - Choose projects with rich metadata
- โจ๏ธ Keyboard Shortcuts - Quick access with Ctrl+Shift+P
- ๐ Military-Grade Encryption - AES-256 encryption for sensitive variables
- ๐ Session Management - Time-limited authentication with configurable TTL
- ๐ Audit Logging - Complete trail of all secure variable operations
- ๐ข Enterprise Features - Secure storage, access controls, and compliance ready
- ๐ ๏ธ CLI Tools - Full command-line interface for all features
- ๐ REST API - Integrate with any tool or workflow
- ๐จ Customizable - Themes, settings, and extensible architecture
- ๐ฅ Team Support - Shared prompts and project configurations
Transform your AI workflow in under 5 minutes! Choose your preferred setup method:
# 1. Clone and install
git clone https://github.com/your-username/promptManager.git
cd promptManager
make install
# 2. Start API server
prompt-api &
# 3. Add your first prompt
prompt-manager add --manual "Daily Standup" "Help me prepare for standup: 1) What I completed yesterday 2) What I'm working on today 3) Any blockers"
# 4. Use it instantly
prompt-manager copy 1 # โ
Copied to clipboard!# Clone the repository
git clone https://github.com/your-username/promptManager.git
cd promptManager
# Production installation
make install
# Or development installation (includes dev tools)
make developAvailable Commands After Installation:
prompt-manager(or short aliaspm) - Main CLI interfacepmcli- Task-Master integrationprompt-api- API serverproject-registry- Project management
# If you prefer the original script approach
pip install flask flask-cors requests numpy
chmod +x prompt_manager.py pmcli setup_integration.sh
./setup_integration.shStep 1: Start API Server
# Navigate to promptManager directory
cd promptManager
# Start the API server (keep this running)
python3 src/prompt_api.pyStep 2: Install Extension in Chrome
- Open Chrome Extensions: Go to
chrome://extensions/ - Enable Developer Mode: Toggle switch in top-right corner
- Load Extension:
- Click "Load unpacked" button
- Navigate to and select:
promptManager/browser_extension/ - Click "Select Folder"
- Pin Extension: Click puzzle piece icon โ pin Prompt Manager
Step 3: Configure Extension
- Right-click extension icon โ "Options"
- Test Connection - Should show โ if API server is running
- Add a project using "Add Manually" button
- Enable desired settings (auto-suggest, show widget, etc.)
Step 4: Test It Out
- Visit Claude.ai
- Focus on text input โ See prompt suggestions appear! โจ
- Press Ctrl+Shift+P โ Quick access popup
- Click ๐ โ Choose Task-Master project for context
# Add your first prompt
prompt-manager add --manual "Code Review" "Review this code for security, performance, and best practices: {code}" -c development
# List your prompts
prompt-manager list
# Copy prompt with variables
prompt-manager copy 1 -v code="function login() { return true; }"
# Get interactive help
prompt-manager help# Add your first project
project-registry add "My Project" ~/path/to/project
# Get instant project context
cd ~/path/to/project
pmcli # โ
Copies: PRD, requirements, tasks, git branch, success criteria- Visit any supported site (Claude.ai, ChatGPT, GitHub, Notion, Linear, Slack)
- Focus text area โ Auto-suggestions appear
- Use shortcuts:
Ctrl+Shift+Pfor quick access - Choose projects with ๐ for context-aware prompts
Store sensitive data like API keys, passwords, and tokens with military-grade security that rivals enterprise password managers.
Encryption Standards:
- AES-256 Encryption - Same standard used by banks and government agencies
- PBKDF2 Key Derivation - 100,000 iterations to prevent brute-force attacks
- System Keychain Integration - Leverages macOS Keychain for password storage
- Secure Memory Handling - Values cleared immediately after use
Access Control:
- Session-based Authentication - Time-limited access (1-hour default)
- Write-only Storage - Values cannot be retrieved in plaintext after creation
- Runtime-only Decryption - Secrets only decrypted when actually used in prompts
- Audit Trail - Complete logging of all operations for compliance
# Add sensitive API key (interactive password prompt)
./pmcli svar add openai_key "OpenAI API Key"
# List all variables (secure ones marked with ๐)
./pmcli var list
# Use in prompts (automatic authentication if needed)
./pmcli copy 1 # Secure variables automatically filled
# Session management
./pmcli svar session info # Check authentication status
./pmcli svar session logout # End session for securityAPI Integration Prompts:
Test the {endpoint} API:
Authorization: Bearer {api_token} # โ Secure variable
Method: {method}
Expected: {expected_response}
Database Operations:
Connect to production database:
Host: {db_host}
Password: {db_password} # โ Secure variable
Execute: {query}
CI/CD Workflows:
Deploy to {environment}:
GitHub Token: {github_token} # โ Secure variable
Docker Registry: {registry_token} # โ Secure variable
Slack Webhook: {slack_webhook} # โ Secure variable
| Feature | Description | Security Level |
|---|---|---|
| Encryption | AES-256-CBC with random IV/salt | Military Grade |
| Key Derivation | PBKDF2-SHA256, 100K iterations | Bank Standard |
| Password Storage | System keychain or encrypted file | Hardware Secured |
| Session Management | Configurable TTL with auto-expire | Enterprise Standard |
| Audit Logging | All operations logged with timestamps | Compliance Ready |
| Memory Security | Immediate cleanup after use | Government Standard |
# Secure Variable Management
./pmcli svar add <name> "description" # Add with password prompt
./pmcli svar list [--show-values] # List (--show-values is DANGEROUS)
./pmcli svar update <name> [-d "desc"] # Update description/value
./pmcli svar delete <name> [-f] # Delete (force flag available)
# Session Management
./pmcli svar session info # Show authentication status
./pmcli svar session logout # End secure session
./pmcli svar session extend # Extend session time
# Security & Compliance
./pmcli svar audit [-n 50] # View audit log
./pmcli svar export backup.json # Export encrypted backup- Passwords are never stored - only secure hashes
- Values are write-only after creation - cannot be retrieved in plaintext
- Sessions auto-expire after configured time (default: 1 hour)
- All operations are audited for security compliance
๐ Complete Security Documentation:
- Secure Variables Guide - User guide and quick reference
- Security Architecture - Technical implementation details and compliance information
Clean up verbose logs before using them in AI prompts - automatically filters repetitive patterns while preserving important information.
# Your clipboard has 500 lines of repetitive logs
./pmcli sanitize
โ
Sanitized content copied to clipboard!
Reduced from 500 to 45 lines (91% reduction)- Intelligent Pattern Detection - Finds similar lines even with slight variations
- Smart Summarization - Replaces repetitive blocks with concise summaries
- Sensitive Data Warning - Alerts you to potential API keys or passwords
- Flexible I/O - Works with clipboard, files, or piped input
# Docker logs
docker-compose logs | ./pmcli sanitize
# Server logs with custom threshold
./pmcli sanitize -i server.log -t 0.8 -s
# Preview without modifying
./pmcli sanitize -p๐ Full Log Sanitization Guide - Complete documentation with examples and best practices
# CLI help - available anytime
prompt-manager help # Interactive help system
pmcli --help # Task-Master integration help
project-registry --help # Project management help
# Start here
prompt-manager help quick-start # 5-minute tutorial- ๐ฏ Quick Start Tutorial - Get up and running in 5 minutes
- ๐ Secure Variables Guide - Enterprise security for sensitive data
- ๐งน Log Sanitization Guide - Clean up verbose logs for AI prompts
- ๐ฎ CLI Command Reference - Every command explained with examples
- ๐ Browser Extension Guide - Installation, usage, and tips
- ๐ง Task-Master Integration - Project context automation
- ๐ Multi-Project Management - Handle multiple projects like a pro
- ๐๏ธ Project Management - Manual project control and settings
- ๐ Security Architecture - Technical implementation, compliance, and threat model
- ๐ค API Reference - REST API for custom integrations
- โ๏ธ Configuration Guide - Settings, environment variables, and customization
- ๐ผ Team Workflows - Multi-user scenarios and best practices
- ๐ข Enterprise Setup - Large organization deployment
- ๐จ Custom Integration - Building your own integrations
# Morning routine: Get project context instantly
cd ~/projects/my-app
pmcli # Copies full project context for AI chat
# Code review: Use standardized review prompts
prompt-manager copy 1 # "Code Review Request" prompt
# Bug investigation: Context-aware debugging prompts
# Extension auto-suggests debugging prompts when "error" detected# Team lead: Set up shared project contexts
project-registry add "Team Project" /shared/team-project
project-registry add "Client Work" /shared/client-work
# Developers: Consistent project context across team
pmcli # Everyone gets same PRD locations, requirements, success criteria- Blog Writing: Prompts for different article types and audiences
- Social Media: Platform-specific content generation prompts
- Documentation: Technical writing and API documentation prompts
- Marketing: Campaign ideas, ad copy, and engagement prompts
# Interactive help - start here!
prompt-manager help
# Available help topics:
prompt-manager help quick-start # 5-minute getting started
prompt-manager help commands # All CLI commands
prompt-manager help task-master # Task-Master integration
prompt-manager help browser # Browser extension setup
prompt-manager help examples # Common use cases
prompt-manager help troubleshooting # Fix common issues- Right-click extension icon โ "Options" โ "Help" tab
- Built-in tutorials and troubleshooting guides
- Connection testing and diagnostics
- Visual setup wizards for common tasks
- ๐ Full Documentation: Browse the docs/ directory
- ๐ฌ Issues: GitHub Issues
- ๐ Bug Reports: Use our issue templates
- ๐ก Feature Requests: We love new ideas!
promptManager/
โโโ ๐ README.md # You are here
โโโ ๐ SECURE_VARIABLES.md # Security documentation
โโโ ๐ง prompt_manager.py # Core CLI tool
โโโ โก pmcli # Task-Master integration + secure variables
โโโ ๐ src/
โ โโโ ๐ prompt_api.py # REST API server
โ โโโ ๐ project_registry.py # Multi-project management
โ โโโ ๐ task_master_*.py # Task-Master integration
โ โโโ ๐ secure_session.py # Session management
โ โโโ ๐ secure_crypto.py # AES-256 encryption
โ โโโ ๐ secure_variables.py # Secure variable manager
โ โโโ ๐งน log_sanitizer.py # Log pattern filtering
โ โโโ ๐ง opus_reasoning.py # AI-powered optimization
โโโ ๐ browser_extension/ # Chrome/Firefox extension
โ โโโ manifest.json
โ โโโ content.js # Page integration magic
โ โโโ options.html # Settings page
โ โโโ popup.html # Extension popup
โโโ ๐ docs/ # Complete documentation
โ โโโ QUICK_START.md # Start here
โ โโโ CLI_USAGE.md # Command reference
โ โโโ BROWSER_EXTENSION.md # Extension guide
โ โโโ TASK_MASTER_INTEGRATION.md
โ โโโ MULTI_PROJECT_GUIDE.md
โ โโโ PROJECT_MANAGEMENT_GUIDE.md
โ โโโ examples/ # Real-world examples
โโโ ๐พ prompts.json # Your prompt database
โโโ ๐ prompts_secure.json # Encrypted secure variables (auto-created)
โโโ โ๏ธ setup_integration.sh # Easy setup script
# Setup takes 30 seconds
make install
# Add a project in 5 seconds
project-registry add "My Project" ~/projects/my-app
# Get perfect AI context in 1 command
pmcli
# โ
Copies: PRD location, requirements, success criteria, current task, git branch
# Paste into Claude Code โ Perfect context every time! - Install extension (2 clicks)
- Visit Claude.ai
- Focus text area โ See prompt suggestions appear ๐ซ
- Click ๐ โ Choose your project
- Click suggestion โ Perfect prompt with full project context!
Just like Grammarly, but for AI prompts!
# Prompt Management
prompt-manager help # Interactive help system
prompt-manager add --manual "Title" "Content" # Add prompt
prompt-manager list # List all prompts
prompt-manager copy 1 # Copy prompt to clipboard
# Secure Variables (NEW!)
pmcli svar add api_key "API Key Description" # Add secure variable
pmcli svar list # List secure variables (๐ marked)
pmcli var list # List all variables (regular + secure)
pmcli svar session info # Check authentication status
# Log Sanitization (NEW!)
pmcli sanitize # Clean clipboard logs (removes repetition)
pmcli sanitize -i app.log -o clean.log # Sanitize log file
pmcli sanitize -p -s # Preview with statistics
# Task-Master Integration
pmcli # Auto-detect project and copy context
pmcli --list-projects # Show available projects
pmcli --select-project # Choose project interactively
# Project Management
project-registry add "Name" /path # Add project manually
project-registry list # List all projects
project-registry disable project-id # Hide project temporarily
# API Server
prompt-api # Start REST API for browser extension- Ctrl+Shift+P - Quick access popup
- Click ๐ - Choose project for Task-Master prompts
- Right-click icon โ "Options" - Full settings and project management
We welcome contributions! Here's how to get started:
- Fork the repository
- Create a feature branch:
git checkout -b feature/amazing-feature - Make your changes and add tests
- Run the test suite:
python -m pytest - Submit a pull request
# Install in development mode with all dev tools
make develop
# Run tests
make test
# Format code
make format
# Start development API server
prompt-api --debug
# Load extension in Chrome developer mode
# Point to browser_extension/ directory- Opus-powered prompt optimization
- Success rate learning and improvement
- Automatic prompt generation from usage patterns
- Shared prompt libraries
- Team analytics and insights
- Role-based access control
- VS Code extension
- Slack bot integration
- API webhooks and automation
- SSO authentication
- Audit logs and compliance
- Advanced analytics dashboard
CLI Tools - Configure via environment variable or command-line flag:
# Environment variable (recommended)
export PROMPT_API_URL="http://localhost:5000/api"
# Or use command-line flag for individual commands
prompt-manager --api-url "http://your-server:5000/api" list
pmcli --api-url "http://your-server:5000/api"Browser Extension - Configure via options page:
- Right-click extension icon โ "Options"
- Update "API URL" field
- Click "Test Connection" to verify
# API Configuration
export PROMPT_API_URL="http://localhost:5000/api"
export PROMPT_MANAGER_DIR="/path/to/promptManager"
# Secure Variables Configuration
export PM_SECURE_SESSION_TTL=60 # Session timeout in minutes
export PM_SECURE_ENCRYPTION_BACKEND="auto" # Encryption backend preference
# Task-Master Integration
export TASK_MASTER_DIR="/path/to/task-master"
# AI Service Keys (DEPRECATED - Use Secure Variables Instead!)
# Migrate these to secure variables for better security:
# ./pmcli svar add anthropic_api_key "Anthropic API Key"
# ./pmcli svar add openrouter_api_key "OpenRouter API Key"
export ANTHROPIC_API_KEY="your-key-here"
export OPENROUTER_API_KEY="your-key-here"# Add to ~/.bashrc or ~/.zshrc for power-user shortcuts
source ~/promptManager/src/claude_integration.sh
# Quick commands now available:
# pms - prompt suggest pmu - prompt use
# pml - prompt list pmf - prompt find
# tm - task-master promptThis project is licensed under the MIT License - see the LICENSE file for details.
- Inspired by tools like Grammarly for seamless user experience
- Built for the Claude Code and Task-Master workflow
- Thanks to the open-source community for the amazing libraries
- ๐ Quick Start - Get running in 5 minutes
- ๐ฎ Try the CLI:
prompt-manager help quick-start - ๐ Install Browser Extension - Like Grammarly for AI prompts
- ๐ Explore Full Documentation - Become a power user
Join thousands of developers, writers, and teams who use Prompt Manager to work smarter with AI.
๐ Get Started Now | ๐ Full Docs | ๐ Browser Extension | ๐ป API Reference