write_struct

Transform raw text into structured, formatted content for WriteFreely and beyond.

write_struct is a Python package designed to enhance content creation by automatically converting user-submitted text into well-structured, formatted output. It leverages AI-powered pattern matching to ensure consistency, readability, and publication-ready quality for articles, blog posts, social media updates, and more.

🚀 Features

• Automated text structuring using AI-driven pattern matching.
• Flexible LLM integration – works with default ChatLLM7 or any LangChain-compatible model.
• Regex-based validation to enforce output consistency.
• Lightweight and efficient for text-based content processing.
• No multimedia handling – focuses purely on text formatting.

📦 Installation

Install via pip:

pip install write_struct

🔧 Usage

Basic Usage (Default LLM7)

from write_struct import write_struct

user_input = "Your raw text here..."
response = write_struct(user_input)
print(response)

Custom LLM Integration

You can replace the default ChatLLM7 with any LangChain-compatible model (e.g., OpenAI, Anthropic, Google Generative AI):

Using OpenAI:

from langchain_openai import ChatOpenAI
from write_struct import write_struct

llm = ChatOpenAI()
response = write_struct(user_input, llm=llm)

Using Anthropic:

from langchain_anthropic import ChatAnthropic
from write_struct import write_struct

llm = ChatAnthropic()
response = write_struct(user_input, llm=llm)

Using Google Generative AI:

from langchain_google_genai import ChatGoogleGenerativeAI
from write_struct import write_struct

llm = ChatGoogleGenerativeAI()
response = write_struct(user_input, llm=llm)

🔑 API Key Configuration

• Default: Uses LLM7_API_KEY from environment variables.
• Manual Override: Pass the API key directly:

  response = write_struct(user_input, api_key="your_api_key_here")
  

• Free API Key: Register at https://token.llm7.io/ for LLM7.

📜 Input Parameters

📊 Rate Limits

• LLM7 Free Tier: Sufficient for most use cases.
• Upgrade: For higher limits, use a custom API key or switch to a paid plan.

📝 License

MIT License.

📢 Support & Issues

For bugs, feature requests, or support, open an issue on GitHub.

👤 Author

Eugene Evstafev (@chigwell) 📧 hi@euegne.plus