promptcoder-cli

PromptCoder CLI

🤖 AI-powered code generation CLI tool with conversation persistence, advanced file tools, and support for multiple LLM providers (OpenAI GPT & Anthropic Claude).

✨ Features

• 🎯 Interactive Mode: Chat-like interface for iterative code generation
• 📝 Conversation Persistence: Save and resume conversations across sessions
• 🔧 Advanced File Tools: Diff edits, search, checkpoints, and smart directory filtering
• 🤖 Multiple LLM Providers: OpenAI (GPT-4) and Anthropic (Claude) support
• 🚀 Retry Logic: Automatic retry with exponential backoff for API rate limits
• 🎨 Beautiful Interface: Colorized output with progress indicators
• 📁 Project-Aware: Maintains working directory context per conversation

🚀 Installation

Global Installation (Recommended)

npm install -g promptcoder-cli

Local Installation

npm install promptcoder-cli

⚙️ Configuration

First Run Setup

promptcoder config

Environment Variables

# OpenAI
export OPENAI_API_KEY="your-openai-api-key"
export OPENAI_MODEL="gpt-4o"  # optional

# Anthropic
export ANTHROPIC_API_KEY="your-anthropic-api-key"
export ANTHROPIC_MODEL="claude-3-5-sonnet-20241022"  # optional

Manual Configuration

Configuration is stored in ~/.promptcoder/config.json

🎯 Usage

Interactive Mode

promptcoder interactive
# or
promptcoder i

Single Prompt

promptcoder prompt "Create a React component for a todo list"
# or
promptcoder p "Add error handling to my express server" --directory ./my-project

Conversation Management

# In interactive mode:
save          # Save current conversation
load          # Load a previous conversation
list          # List all saved conversations
rename        # Rename current conversation
delete        # Delete a conversation
clear         # Clear current conversation

🔧 Available Tools

The AI has access to these powerful tools:

Basic File Operations

• read_file - Read file contents
• write_file - Write content to files (creates directories automatically)
• read_directory - Smart directory listing (filters node_modules, .git, etc.)
• create_directory - Create directories
• delete_file - Delete files
• file_exists - Check if files/directories exist

Advanced Editing

• edit_file_diff - Precise edits using exact text matching
• insert_lines - Insert content at specific line numbers
• delete_lines - Remove specific line ranges

Search & Discovery

• search_in_files - Search patterns across multiple files with regex support

Version Control

• create_checkpoint - Save snapshots of file states
• list_checkpoints - View all checkpoints with metadata
• restore_checkpoint - Rollback to previous states
• show_file_diff - Compare current files with checkpoint versions

📚 Examples

Creating a React App

promptcoder i
Prompt: Create a modern React TypeScript app with routing and state management

# The AI will:
# 1. Read your directory structure
# 2. Create package.json with dependencies
# 3. Set up TypeScript configuration
# 4. Create components and routing
# 5. Set up state management
# 6. Add proper file structure

Refactoring Code

promptcoder i
Prompt: Refactor this codebase to use TypeScript and add proper error handling

# The AI will:
# 1. Create a checkpoint of current state
# 2. Search through existing files
# 3. Convert JavaScript to TypeScript
# 4. Add error handling patterns
# 5. Update configurations
# 6. Show diffs of changes made

Conversation Persistence

# Start working on a project
promptcoder i
Prompt: Build a REST API with Express and PostgreSQL
# ... work continues ...
save
Name: "E-commerce API"
Description: "Building product catalog and user auth"

# Later, resume exactly where you left off
promptcoder i
load
# Select: "E-commerce API (15 messages, 12/5/2024)"
# Continue with full context preserved

🔄 Conversation Flow

🤖 PromptCoder Interactive Mode
Commands: exit, clear, save, load, list, rename, delete

[My React Project] Prompt: Add authentication to the app

🔄 Processing (1/25)...

💬 Response:
I'll help you add authentication to your React app. Let me first examine your current project structure.

🔧 Executing 2 tool call(s):
  → read_directory({"dir_path":"."})
    Contents of . (8 items):
    FILE: package.json (2KB)
    DIR: src
    FILE: tsconfig.json (421B)
    ...

  → read_file({"file_path":"src/App.tsx"})
    File contents of src/App.tsx:
    ...

🔄 Continuing conversation after tool execution...

💬 Response:
Now I'll create the authentication system. I'll add:
1. Auth context and provider
2. Login/signup components
3. Protected routes
4. JWT token management

🔧 Executing 4 tool call(s):
  → create_checkpoint({"checkpoint_name":"before_auth","description":"Before adding authentication"})
    Created checkpoint "before_auth" with 12 files

  → write_file({"file_path":"src/contexts/AuthContext.tsx","content":"..."})
    Successfully wrote 1240 characters to src/contexts/AuthContext.tsx
  ...

✅ Turn completed

🛠️ Development

# Clone and install
git clone https://github.com/promptcoder/promptcoder-cli.git
cd promptcoder-cli
npm install

# Development mode
npm run dev

# Build
npm run build

# Test global installation
npm run build && npm link

🔒 Privacy & Security

• API Keys: Stored locally in ~/.promptcoder/config.json
• Conversations: Stored locally in ~/.promptcoder/conversations/
• No Data Sharing: All data remains on your machine
• Secure: Uses official OpenAI and Anthropic SDKs

🚨 Requirements

• Node.js: 16.0.0 or higher
• API Key: OpenAI or Anthropic API key
• Operating System: macOS, Linux, or Windows

📄 License

MIT © PromptCoder Team

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

📞 Support

• Issues: GitHub Issues
• Discussions: GitHub Discussions

Made with ❤️ for developers who want to code faster with AI assistance.