devibe

Remove auto-generated comments from your codebase. A simple, extensible library for cleaning up AI-generated and IDE-generated comments.

Installation

PyPI (Coming Soon!)

🚀 PyPI package coming soon! We're preparing to publish devibe to the Python Package Index. Once available, you'll be able to install it with:

# Coming soon!
pip install devibe

Once published, you'll be able to run devibe ./myfile.py directly from anywhere.

Local Development Install

Clone the repository and install in development mode:

git clone https://github.com/nhunter0/devibe.git
cd devibe
pip install -e .

After installation, you can run devibe either:

1. As a Python module (works anywhere):

   python -m devibe myfile.py
   

2. Direct command (if pip Scripts directory is in PATH):

   devibe myfile.py
   

Usage

Command Line

Clean a single file:

devibe myfile.py

Clean a directory recursively:

devibe ./src -r

Dry run to see exactly what lines would be removed:

devibe myfile.py --dry-run

This will show output like:

myfile.py - Would remove 3 line(s):

  1. # Generated by Claude
  2.     x = 1  # ✅ AI-generated
  3. # ... rest of implementation

Process only specific file types:

devibe ./src -r -e .py .js .java

Python API

from devibe import CommentRemover

remover = CommentRemover()

# Clean a single file
removed_count, removed_comments = remover.clean_file("myfile.py")

# Clean a directory
stats = remover.clean_directory("./src", recursive=True, extensions=[".py", ".js"])
print(f"Removed {stats['comments_removed']} comments from {stats['files_modified']} files")

Default Rules

devibe comes with intelligent rules to detect and remove:

AI-Generated Comments

• generated by Claude, AI-generated, Created by GPT
• Comments from Copilot, Codeium, TabNine, and other AI assistants

IDE-Generated Comments

• Auto-generated code stubs and boilerplate
• IDE template comments from VSCode, IntelliJ, Eclipse

Status Markers & Emojis

• Emoji indicators (✅, 🔒, 🚀, etc.)
• Status prefixes (FIXED:, REMOVED:, ENHANCED:, SIMPLIFIED:)
• Case-insensitive matching with or without colons

Placeholder Comments

• ... rest of code, ... existing methods
• your code here, insert logic here
• TODO: implement, stub, placeholder

List active rules:

devibe --list-rules

Adding Custom Rules

I have definetly missed rules so if you have any feel free to go to: /devibe/rules.py add a custom rule send a PR.

Via Command Line

Add custom patterns on the fly:

devibe myfile.py --add-pattern "my_rule" "pattern_to_match"

Via Python

from devibe import CommentRemover, PatternRule

remover = CommentRemover()

# Add a custom pattern rule
custom_rule = PatternRule(
    name="custom_assistant",
    description="Remove specific assistant comments",
    patterns=[r"generated by my-assistant", r"my-assistant wrote this"]
)
remover.rule_registry.add_rule(custom_rule)

# Now clean files with your custom rule active
remover.clean_file("myfile.py")

Creating Your Own Rule Class

from devibe.rules import Rule

class KeywordRule(Rule):
    def __init__(self, keywords):
        super().__init__("keyword_rule", "Remove comments with specific keywords")
        self.keywords = [k.lower() for k in keywords]

    def matches(self, comment):
        comment_lower = comment.lower()
        return any(keyword in comment_lower for keyword in self.keywords)

# Use your custom rule
remover = CommentRemover()
remover.rule_registry.add_rule(KeywordRule(["autogen", "placeholder"]))

Requirements

• Python 3.7+
• No external dependencies for core functionality

Development

Install in development mode with:

git clone https://github.com/nhunter0/devibe.git
cd devibe
pip install -e .

Run tests:

pytest tests/

Contributing

We welcome contributions! To add support for new languages or improve detection rules:

1. Fork the repository
2. Create a feature branch (git checkout -b feature/amazing-feature)
3. Add your parser or rule to the appropriate registry
4. Write tests for your changes
5. Submit a pull request

Roadmap

• PyPI package publication
• Devibe UI & UX !
• VS Code extension
• More language-specific rules

License

MIT

Author

Created with ❤️ to keep codebases clean and professional.

---