Repository2Prompt is a powerful tool that converts GitHub repositories or local directories into prompts for Large Language Models (LLMs). It extracts the structure and content of repositories, generating structured output for easy understanding and processing by LLMs.
- Support for both GitHub repositories and local directories
- Generation of file tree structures
- Extraction of specified file contents
- Multiple output formats: Markdown, JSON, plain text, and split format
- Customizable template system
- Flexible configuration options
You can install Repository2Prompt using pip:
pip install repository2promptAlternatively, you can install from source:
git clone https://github.com/wctsai20002/repository2prompt.git
cd repository2prompt
pip install .Repository2Prompt provides a simple command-line interface:
repository2prompt [OPTIONS] INPUT_PATHOptions:
-t, --template PATH: Path to a custom Jinja2 template file-f, --format [markdown|json|text|split]: Output format (default: markdown)-o, --output PATH: Output file path (default: repository_name.{format})--help: Show help message
Example:
# Convert a GitHub repository to JSON format and save to a specific file
repository2prompt https://github.com/octocat/octocat.github.io -f json -o output.json
# Convert a local directory to markdown and save with default name
repository2prompt /path/to/local/repo -f markdown
# Use a custom template and save the output to a specific file
repository2prompt https://github.com/user/repo -t custom_template.j2 -o custom_output.mdIf no output file is specified, the result will be saved in the current directory with the name repository_name_prompt.{extension}, where {extension} is the chosen output format.
You can also use Repository2Prompt in your Python code:
from repository2prompt import Repository2Prompt
converter = Repository2Prompt("https://github.com/octocat/octocat.github.io", output_format="json")
result = converter.process()
print(result)Generates prompts in split format, and then sends each prompt to LLM sequentially.
import yaml
import json
from repository2prompt import Repository2Prompt, CONFIG
converter = Repository2Prompt("https://github.com/octocat/octocat.github.io", output_format="split")
result = converter.process()
prompts = json.loads(result)
for item in prompts:
prompt = item['prompt']
content = item['content']
# Combine prompt and content
full_message = f"{prompt}\n\n{content}"
print(full_message)
# Send to LLM and get response
response = chat_with_llm(full_message)Repository2Prompt uses a YAML configuration file. The default configuration is located in default_config.yaml within the package. You can override the default settings by creating a .repository2prompt.yaml file in your home directory.
Key configuration options include:
supported_output_formats: List of supported output formatsmax_file_size: Maximum file size to process (in bytes)supported_file_extensions: List of supported file extensionsignore_dirsandignore_files: Lists of directories and files to ignore
from repository2prompt import Repository2Prompt
converter = Repository2Prompt("https://github.com/octocat/octocat.github.io")
result = converter.process()
print(result)from repository2prompt import Repository2Prompt
converter = Repository2Prompt("/path/to/local/repo", output_format="json")
result = converter.process()
print(result)from repository2prompt import Repository2Prompt
converter = Repository2Prompt("https://github.com/octocat/octocat.github.io",
template_path="path/to/custom_template.j2",
output_format="text")
result = converter.process()
print(result)Repository2Prompt uses a YAML configuration file for its settings. There are several ways to customize the configuration:
The default configuration is located in repository2prompt/default_config.yaml within the package.
You can create a .repository2prompt.yaml file in your home directory to override the default settings. This file will be automatically loaded if it exists.
For project-specific settings, you can create a custom YAML file and load it in your code. For example:
import yaml
from repository2prompt import Repository2Prompt, CONFIG
# Load custom configuration
with open('path/to/your/custom_config.yaml', 'r') as f:
custom_config = yaml.safe_load(f)
# Update the default configuration
CONFIG.update(custom_config)
# Use the updated configuration
converter = Repository2Prompt("https://github.com/user/repo")
result = converter.process()You can also specify a custom configuration file using the REPOSITORY2PROMPT_CONFIG environment variable:
export REPOSITORY2PROMPT_CONFIG=/path/to/your/custom_config.yamlKey configuration options include:
github_api_token: Github api tokensupported_output_formats: List of supported output formatsmax_file_size: Maximum file size to process (in bytes)supported_file_extensions: List of supported file extensionsignore_dirsandignore_files: Lists of directories and files to ignoremax_files_to_process: Maximum number of files to processmax_prompt_length: Maximum length of the generated promptsplit_format_prompts: Prompts of output
Example of a custom configuration file:
github_api_token: "xxx_xxxxxxxxxx"
max_file_size: 2097152 # 2MB
ignore_dirs:
- node_modules
- .git
- build
supported_file_extensions:
- .py
- .js
- .md
max_files_to_process: 50Remember that any setting not specified in your custom configuration will use the default value from the package's default configuration.
We welcome contributions of all forms! If you'd like to contribute to Repository2Prompt, please follow these steps:
- Fork the repository
- Create your feature branch (
git checkout -b feature/AmazingFeature) - Commit your changes (
git commit -m 'Add some AmazingFeature') - Push to the branch (
git push origin feature/AmazingFeature) - Open a Pull Request
Before submitting a Pull Request, please ensure your code passes all tests and adheres to our coding style guidelines.
This project is licensed under the MIT License - see the LICENSE file for details.
Project Link: https://github.com/wctsai20002/repository2prompt