ContentForge

CLI tool for generating content using LLMs from your terminal.

Generate blog posts, social media content, emails, ad copy, and more — powered by OpenAI, Gemini, or local Ollama models.

Installation

pip install contentforge

Quick Start

# Configure your API key
contentforge config set openai_api_key sk-your-key-here

# Generate a blog post
contentforge generate blog --topic "AI trends in 2026" --tone professional

# Generate a LinkedIn post
contentforge generate social --platform linkedin --topic "Remote work" --goal engagement

# Generate with streaming (default)
contentforge generate email --subject "Product launch" --type marketing

# Save to file
contentforge generate blog --topic "Python tips" -o blog.md

# Copy to clipboard
contentforge generate social --topic "Coding" --copy

Supported Content Types

Command	Description
generate blog	Blog posts with SEO optimization
generate social	Platform-optimized social media posts
generate email	Professional emails with subject lines
generate tweet-thread	Twitter/X threads
generate ad	Ad copy for Google, Facebook, Instagram, LinkedIn
generate seo	SEO meta tags and keywords
generate product	Product descriptions
generate youtube	YouTube video descriptions

Providers

ContentForge supports three LLM providers:

Provider	Models	Setup
OpenAI	gpt-4o, gpt-4o-mini, gpt-4-turbo	contentforge config set openai_api_key YOUR_KEY
Gemini	gemini-2.0-flash, gemini-1.5-pro	contentforge config set gemini_api_key YOUR_KEY
Ollama	llama3.2, mistral, phi3 (local)	Just have Ollama running locally

# Switch default provider
contentforge config set default_provider gemini

# Use a specific provider for one command
contentforge generate blog --topic "AI" --provider ollama --model mistral

Common Options

All generate commands support these options:

Option	Description
--provider / -p	LLM provider (openai/gemini/ollama)
--model / -m	Model override
--output / -o	Save output to file
--format / -f	Output format (markdown/plain/json)
--copy	Copy result to clipboard
--stream / --no-stream	Enable/disable streaming
--temperature	Sampling temperature (0.0-2.0)
--max-tokens	Maximum output tokens

Configuration

# Show current config
contentforge config

# Set a value
contentforge config set default_provider gemini
contentforge config set default_temperature 0.5

# Show config file path
contentforge config path

Configuration is stored in ~/.contentforge/config.toml. Environment variables override config file values using the CONTENTFORGE_ prefix:

export CONTENTFORGE_OPENAI_API_KEY=sk-your-key
export CONTENTFORGE_DEFAULT_PROVIDER=gemini

Templates

# List all templates
contentforge templates

# Show template details
contentforge templates show blog

Development

git clone https://github.com/brolyroly007/contentforge.git
cd contentforge
make dev        # Install with dev dependencies
make test       # Run tests
make lint       # Run linter
make format     # Auto-format code

License

MIT