anything-to-text 0.3.2

Export source code and project files into AI-friendly text bundles with manifest

anything-to-text

Export source code and project files into AI-friendly text bundles with a compact manifest.

Commands: anything-to-text or att (short alias)

Features

• Multi-language support: Python, TypeScript, JavaScript, Java, Kotlin, Swift, HTML/CSS, and more
• AI-optimized output: Creates bundle files with embedded metadata for AI assistants
• Smart filtering: Respects .gitignore and skips build/cache directories automatically
• Manifest generation: Produces MANIFEST.json with precise file location mappings
• Project overview: Optional PROJECT_OVERVIEW.txt for architecture insights
• Secret protection: Skips .env, keys, certificates, and other sensitive files by default
• Fancy interactive UI: Checkbox-style selection with questionary
• Large file handling: Automatically splits large files into manageable chunks

Installation

pip install anything-to-text

Quick Start

CLI Usage

# Show version
anything-to-text --version
att --version

# Interactive mode - scan current directory
anything-to-text
att

# Specify project root
anything-to-text /path/to/project
att /path/to/project

# Non-interactive mode with specific categories
anything-to-text /path/to/project --non-interactive --categories python typescript
att /path/to/project --non-interactive --categories python typescript

# Export specific files or folders
anything-to-text . --non-interactive --paths src main.py
att . --non-interactive --paths src main.py

# Generate project overview
anything-to-text . --project-overview
att . --project-overview

# Use fancy interactive UI
anything-to-text . --fancy-ui
att . --fancy-ui

# Set custom output directory
anything-to-text . --output-dir ./my-bundles
att . --output-dir ./my-bundles

# Adjust max lines per bundle
anything-to-text . --max-lines 10000
att . --max-lines 10000

# Include secret files (use with caution)
anything-to-text . --include-secret-files
att . --include-secret-files

Python API

from pathlib import Path
from anything_to_text import CodeExporter, ExportConfig

# Simple export
exporter = CodeExporter()
result = exporter.export(
    project_root="/path/to/project",
    categories=["python", "typescript"],
    interactive=False,
    project_overview=True,
)

print(f"Created {result.bundles_created} bundles")
print(f"Exported {result.files_exported} files")
print(f"Output directory: {result.output_dir}")

# Or use configuration object
config = ExportConfig(
    project_root=Path("/path/to/project"),
    max_lines=10000,
    include_secret_files=False,
    categories=["python", "web_ui"],
    interactive=False,
)

exporter = CodeExporter(config)
result = exporter.export()

Output Files

After running, you'll get:

File	Description
bundle_001.txt, bundle_002.txt, ...	Text bundles containing your source code
MANIFEST.json	Maps files to bundles and line ranges
README_EXPORT.txt	Summary of the export
PROJECT_OVERVIEW.txt	(Optional) Architecture overview

Categories

Category	Description	Extensions
python	Python backend / scripts	.py
typescript	TypeScript / TSX	.ts, .tsx
javascript	JavaScript / JSX	.js, .jsx, .mjs, .cjs
java_kotlin	Android / JVM	.java, .kt, .kts, .gradle
ios_apple	iOS / Apple platform	.swift, .m, .h, etc.
web_ui	HTML / CSS / UI	.html, .css, .scss, .vue, etc.
config_docs	Config / metadata	.json, .yaml, .xml, .toml, .md

Category Aliases

Alias	Maps to
py	python
ts, tsx	typescript
js, jsx	javascript
java, kotlin, android	java_kotlin
ios, swift, apple	ios_apple
web, frontend, ui	web_ui
config, docs	config_docs

Usage Examples

Export Python and TypeScript code for AI review

att /my-project --non-interactive --categories python typescript --project-overview

Export specific folders

att . --non-interactive --paths src/components src/utils

Full export with all features

att /path/to/project \
    --non-interactive \
    --categories python typescript web_ui \
    --max-lines 10000 \
    --project-overview \
    --output-dir ./ai-review-bundles

Interactive mode with fancy UI

att . --fancy-ui

Use with AI assistants

After running the exporter:

1. Attach bundles: Upload bundle_001.txt (and others) to your AI assistant
2. Include manifest: Upload MANIFEST.json so the assistant understands file mappings
3. Add overview: Upload PROJECT_OVERVIEW.txt for architecture context
4. Ask questions: The assistant can now reference exact file locations and line numbers

Security

By default, the following files are skipped to protect sensitive data:

• .env files and variants (*.env.*)
• Private keys (*.pem, *.key, id_rsa, id_ed25519)
• Certificates (*.crt, *.cer, *.der)
• Keystores (*.jks, *.keystore, *.p12, *.pfx)
• Mobile provisioning (*.mobileprovision)
• Firebase/Google config (google-services.json, GoogleService-Info.plist)

Use --include-secret-files to override this behavior (use with caution).

CLI Options

usage: anything-to-text [-h] [--max-lines MAX_LINES] [--output-dir OUTPUT_DIR]
                        [--non-interactive] [--categories [CATEGORIES ...]]
                        [--paths [PATHS ...]] [--include-secret-files]
                        [--fancy-ui] [--project-overview] [--version]
                        [project_root]

positional arguments:
  project_root          Project root to scan (default: current directory)

options:
  -h, --help            Show help message
  --version             Show version
  --max-lines N         Max lines per bundle (default: 8000)
  --output-dir DIR      Custom output directory
  --non-interactive     Run without prompts
  --categories CATS     Categories to export (e.g., python typescript)
  --paths PATHS         Specific files/folders to export
  --include-secret-files Include secret-like files
  --fancy-ui            Use checkbox-style interactive UI
  --project-overview    Generate PROJECT_OVERVIEW.txt

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT License - see LICENSE file for details.