Introducing m1f: Feed Your AI the Whole Story

·
Georg Franz, m1f@franz.agency
announcement ai tools

Working with AI assistants like Claude, GPT-4, or other LLMs on coding projects can be frustrating. You paste one file, then another, then realize you need to show a config file, then a test file… Before you know it, you’ve spent more time copying and pasting than actually coding.

The Problem with Traditional AI Context

When working with AI on code projects, developers face several challenges:

  • Fragmented Context: Constantly switching between files to copy relevant code
  • Missing Dependencies: Forgetting to include important related files
  • Token Limits: Manually managing what to include within token constraints
  • Repetitive Work: Re-sharing the same context for follow-up questions

Enter m1f: Make One File

m1f (Make One File) solves these problems by intelligently bundling your entire project into a single, AI-optimized file. It’s like giving your AI assistant x-ray vision into your codebase.

Key Features

  1. Smart Bundling: Automatically includes all relevant files based on your configuration
  2. Token Optimization: Respects AI model token limits while maximizing context
  3. Security First: Built-in filtering for sensitive files and data
  4. Multiple Formats: Supports XML, Markdown, and plain text output
  5. Tool Suite: Includes specialized tools for different use cases

Getting Started

Installation is simple:

git clone https://github.com/franz-agency/m1f.git
cd m1f
./install

Then, in any project directory:

m1f

That’s it! Your entire project is now in .m1f/_bundle.md, ready to paste into your AI assistant.

Real-World Use Cases

1. Code Reviews

Bundle your feature branch changes for comprehensive AI review:

m1f --git-diff main

2. Documentation Generation

Include all source files for accurate documentation:

m1f --preset python-full

3. Debugging Sessions

Share your entire debugging context:

m1f --include "src/**/*.js" --include "**/*.test.js"

The m1f Ecosystem

m1f comes with a suite of specialized tools:

  • s1f: Split files back out from bundles
  • m1f-rag: Create RAG-optimized bundles
  • m1f-web: Bundle web pages and documentation
  • m1f-scrape: Advanced web scraping with AI parsing
  • token-count: Count tokens before sending to AI

What’s Next?

We’re just getting started. Our roadmap includes:

  • IDE integrations for seamless bundling
  • Cloud sync for team collaboration
  • Custom AI model configurations
  • Advanced caching and incremental updates

Join the Community

m1f is open source and we welcome contributions! Check out our GitHub repository to get involved.

Ready to revolutionize how you work with AI? Get started with m1f today!

Related Posts

5 Bundle Strategies for Optimal AI Context

Learn proven strategies for creating effective AI-optimized bundles that maximize context while respecting token limits.

The Irony of Claude Code: When AI Needs to Google Itself

A tale of how a multi-billion dollar AI needs to look up its own documentation online – and how m1f helps solve this existential crisis.

Back to all posts