cmake2chat

Lost in a huge CMake project? Wish an LLM could just "see" the codebase structure and all build rules? Now they can!

What's This?

cmake2chat scans your CMake project directory, builds a tree view of the codebase, and exports the tree view, the paths of all CMakeLists.txt files along with their contents to an OpenAI Chat Completions-compatible JSON.

Typical Use Cases

• Instant overview: Give an LLM a complete picture of your project's structure (files, folders) plus CMake build logic at every level.
• Build-system-powered Q&A: Ask questions like "What libraries does this project build?", "Where is the executable defined?", or "What targets/flags are set in this module?" and get relevant, context-based answers!
• AI onboarding: Drop into a new CMake project and let an LLM guide you using actual high-level structure and build scripts, not just guesses.
• No manual plumbing: Skip hours of clicking/repeating find ., opening dozens of files, or pasting snippets piecemeal. Feed the AI the whole map and let it answer or advise precisely.

Install

pip install cmake2chat

Example

Suppose you have the following CMake project directory structure:

- my_project/
  - CMakeLists.txt
  - src/
    - CMakeLists.txt
    - main.cpp
  - lib/
    - CMakeLists.txt

To generate an OpenAI Chat Completions-compatible JSON describing your project, simply run:

python -m cmake2chat my_project -o my_project.json

An example of the beginning of the output might look like:

[
  {
    "role": "user",
    "content": "- my_project/\n  - CMakeLists.txt\n  - src/\n    - CMakeLists.txt\n    - main.cpp\n  - lib/\n    - CMakeLists.txt"
  },
  {
    "role": "user",
    "content": "CMakeLists.txt"
  },
  {
    "role": "user",
    "content": "# Top-level build logic goes here\ncmake_minimum_required(VERSION 3.10)\nproject(MyProject)\n..."
  },
  {
    "role": "user",
    "content": "src/CMakeLists.txt"
  },
  {
    "role": "user",
    "content": "# Build logic for src directory\nadd_executable(main main.cpp)\n..."
  }
]

You can now use this JSON directly as input to an LLM API for a powerful, context-aware understanding of your CMake project!

How it works

• Scan: Recursively walks your project directory.
• Map: Extracts the structure (files & folders) as a tree.
• Collect: For each CMakeLists.txt, grabs its path and full content.
• Export: Outputs everything as an OpenAI Chat Completions-compatible JSON.

Contributing

Contributions are welcome! Please submit pull requests or open issues on the GitHub repository.

License

This project is licensed under the MIT License.