CodeExportForAI

Tool to export any project folder or repository into a single, neatly formatted file — ideal for quick AI-assisted code review, debugging and refactoring. The script collects source files recursively, wraps each file in fenced code blocks with relative paths, filters common noise (e.g. node_modules, .git, images), and produces a paste-ready output or copies it to the clipboard.

Why use CodeExportForAI

• Prepare full project context for AI quickly: paste the entire codebase into ChatGPT/Claude without manual copying.
• Fast code review and debugging: get a consolidated snapshot to ask focused questions about structure, bugs or refactoring.
• Share reproducible context: include relative paths and file order so AI or reviewers can follow the codebase layout.
• Filter and reduce noise: automatically ignore large binaries, images and common vendor folders to keep the export relevant.

Typical use cases

• Code review and refactoring requests to AI assistants (send whole project context in one paste).
• Pair-programming / debugging sessions where you need to show multiple files at once.
• Student help and tutoring: submit a full assignment for constructive feedback.
• Quick repository snapshots for onboarding, audits or issue reproduction.

Features

• Recursively scans a directory and collects source files
• Configurable ignore rules for directories, filenames and extensions (config.py)
• Save output to a file and/or copy to the clipboard
• Print simple statistics (file count, characters, runtime)
• Works with CLI or GUI folder picker (Tkinter)

Installation / Requirements

• Requires Python 3.6+ (recommended)

• Optional clipboard support: install pyperclip (recommended for macOS/Linux or for a more reliable cross-platform clipboard).

Install optional dependency (one of the options below):

# install only pyperclip
pip install pyperclip

# or install from project's requirements.txt
pip install -r requirements.txt

Quickstart

1. Open a terminal in the CodeExportForAI folder.
2. Run:

python code_export_for_AI.py

Or provide a folder and output file directly:

python code_export_for_AI.py -d "C:\path\to\project" -o export.txt

3. If you run without -d, a folder selection dialog opens. The script will create output.txt by default and may copy the content to the clipboard if enabled. Clipboard support is cross-platform: the tool uses pyperclip when available, otherwise falls back to native utilities (clip, pbcopy, xclip/xsel).

Sample output

Each file is exported with a relative path header followed by a fenced code block. Example:

src/main.py:
```python
def hello():
    print("Hello World")
```

components/button.js:
```javascript
function Button() {
    return <button>Click me</button>
}
```

Configuration

The script loads config.py from the same folder (if present). Defaults are used otherwise. Key options:

• BLACKLIST_EXTENSIONS — set of file extensions to ignore
• BLACKLIST_DIRS — directories to skip (e.g. node_modules, .git)
• OUTPUT_FILENAME — default output file name
• MAX_FILE_SIZE_MB — maximum file size to include
• CREATE_FILE — whether to write the output file
• COPY_TO_CLIPBOARD — whether to copy result to clipboard
• BLACKLIST_FILENAMES — filenames to ignore
• FILENAME_FILTER_MODE — 'exact' or 'contains'
• FILENAME_FILTER_MODE — 'exact' or 'contains'
• USE_PYGMENTS — (bool) enable pygments-based detection of fenced-code language tags (default: True).
• EXTENSION_LANGUAGE_MAP — dict mapping extensions (no dot) to language tags used in fenced code blocks. Used as a fallback and fully user-overridable.

When USE_PYGMENTS is enabled and pygments is installed the script will try to auto-detect language aliases from filename+content; otherwise it falls back to EXTENSION_LANGUAGE_MAP. If no language is found the code fence remains untagged. To customize, edit config.py.

Tips

• For large repositories, increase MAX_FILE_SIZE_MB or run on a subset of folders.
• If you rely on clipboard copying on Linux, ensure xclip or xsel is installed or install pyperclip.

Contributing

Improvements welcome — open an issue or submit a pull request.