Skip to content

A powerful Go-based CLI tool that processes text files through Claude AI with different roles and prompts.

Notifications You must be signed in to change notification settings

jameslaneovermind/claude-copywriter

Repository files navigation

Claude Copywriting System

A powerful Go-based CLI tool that processes text files through Claude AI with different roles and research capabilities.

Features

  • 🎭 Multiple Roles: Copywriter, Researcher, Reviewer, Formatter, Summarizer, Editor, and more
  • 🔍 AI Research: Deep research integration via Perplexity API with real-time data and citations
  • 📁 Batch Processing: Process entire directories of files
  • 🚀 Concurrent Processing: Configurable concurrent API calls
  • 📖 Shared Content: Include brand guidelines, company info, and other context
  • 🔒 Privacy-Aware: Sensitive data directories excluded from version control

Quick Start

  1. Install and build:

    git clone <your-repo>
    cd claude-copywriter
    go build -o claude-copywriter
  2. Set API keys:

    export ANTHROPIC_API_KEY='your-api-key-here'
    export PERPLEXITY_API_KEY='your-perplexity-key'  # Optional, for research
  3. Basic usage:

    ./claude-copywriter -input document.md -role copywriter

Directory Structure

claude-copywriter/
├── main.go, client.go, processor.go    # Core application
├── config.json                         # Configuration and roles
├── shared/                              # Brand content (gitignored)
├── inputs/                              # Input files (gitignored)
├── shared-example/                      # Example structures
└── tests/                               # Test files

Core Usage

Basic Processing

# Transform content with different roles
./claude-copywriter -input article.md -role copywriter
./claude-copywriter -input docs.md -role technical-writer
./claude-copywriter -input content.md -role editor

Research Integration

# Research-only mode (saves research to file)
./claude-copywriter -input article.md -research "DevOps trends 2024" -research-only

# Research + processing (includes research in context)
./claude-copywriter -input article.md -role copywriter -research "cloud security trends"

Shared Content

# Include brand guidelines and context
./claude-copywriter -input content.md -role copywriter -include "brand-voice,company-info"

# Include all shared content
./claude-copywriter -input content.md -role copywriter -include "*"

# Pattern matching
./claude-copywriter -input content.md -role copywriter -include "brand"

Batch Processing

# Process all files in a directory
./claude-copywriter -batch inputs/ -role copywriter -concurrent 3

Research Capabilities

The Perplexity API integration provides:

  • Real-time information: Current web content and trends
  • Authoritative sources: Citations from reliable publications
  • Technical depth: Latest API docs and best practices

Example Research Workflow

# 1. Research current trends
./claude-copywriter -input comparison.md -research "infrastructure scanning tools 2024" -research-only

# 2. Create enhanced content with research
./claude-copywriter -input comparison.md -role copywriter -research "TFLint vs security tools" -include "brand-voice"

# 3. Technical deep dive
./claude-copywriter -input comparison.md -role technical-writer -research "Terraform security best practices"

Configuration

Available Roles

./claude-copywriter -list-roles

Built-in roles include: copywriter, researcher, reviewer, formatter, summarizer, editor, seo-optimizer, social-media, technical-writer, blog-writer, email-marketer, academic-writer, creative-writer, analyst, translator, natural.

Custom Configuration

Edit config.json to:

  • Add custom roles and prompts
  • Configure API endpoints and models
  • Set default parameters

Shared Content Setup

# List available shared content
./claude-copywriter -list-shared

# Create shared content files
mkdir -p shared
echo "# Brand Voice Guidelines..." > shared/brand-voice.md
echo "# Company Information..." > shared/company-info.md

Security & Privacy

🔒 Important: shared/ and inputs/ directories are excluded from version control by default.

Best Practices:

  • Keep sensitive content in excluded directories
  • Use environment variables for API keys
  • Never commit real API keys to version control

API Key Setup

Claude API (Required)

  1. Get API key from Anthropic Console
  2. Set: export ANTHROPIC_API_KEY='your-key'

Perplexity API (Optional - for research)

  1. Get API key from Perplexity AI
  2. Set: export PERPLEXITY_API_KEY='your-key'

Advanced Usage

Custom Prompts

./claude-copywriter -input content.md -role copywriter -prompt "Focus on technical accuracy and include code examples"

Output Control

./claude-copywriter -input article.md -role copywriter -output custom-name.md

Verbose Mode

./claude-copywriter -input content.md -role copywriter -verbose

Common Use Cases

  • Content Enhancement: Add current trends and data to existing articles
  • Technical Documentation: Create comprehensive guides with latest best practices
  • Marketing Copy: Transform content with brand voice consistency
  • Research Reports: Gather and synthesize information from multiple sources
  • Competitive Analysis: Research industry trends and positioning
  • Blog Posts: Create SEO-optimized content with proper structure

Troubleshooting

  • API Errors: Ensure API keys are set correctly
  • File Not Found: Check input file paths and permissions
  • Research Failures: Verify Perplexity API key and network connectivity
  • Build Issues: Ensure Go 1.21+ is installed

For more help: ./claude-copywriter -h

About

A powerful Go-based CLI tool that processes text files through Claude AI with different roles and prompts.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages