Skip to main content

A universal CLI utility to configure AI rules files for any project.

Project description

rules4

A universal CLI utility to configure AI rules files (e.g., .roo/rules, CLAUDE.md, .cursor/rules) for any project, based on the latest industry best practices via live Perplexity research.

Features

  • Supports any language or framework via --lang and --tags options
  • Configures rules for tools like Cursor, Roo, Claude, and more
  • Uses live Perplexity API for up-to-date best practices
  • Dry-run mode to preview changes
  • Prompts before overwriting existing files
  • Simple one-command install (packaged for PyPI)
  • Designed for future MCP integration

Installation

pip install rules4

Usage

Basic Rule Generation

To generate rules for a specific tool (e.g., copilot) for a given language and tags:

rules4 copilot --lang python --tags "pytest,langgraph"

This command will:

  • Use gpt-4-turbo as the primary model (default).
  • Generate rules for Python projects, focusing on pytest and langgraph.
  • Save the rules to .github/copilot-python-pytest,langgraph.md (or similar, depending on the tool).

Advanced Usage

You can specify a primary model, a review model, and enable research:

rules4 copilot --primary gpt4.1 --review claude-4-sonnet --research --lang javascript --tags "react,typescript"

This command will:

  • Use gpt4.1 as the primary model for rule generation.
  • Perform research using Perplexity AI before generating rules.
  • Have claude-4-sonnet review and refine the generated rules.
  • Generate rules for JavaScript projects, focusing on react and typescript.

Generating Rules for All Configured Tools

If you have a .rules4rc file configured, you can generate rules for all specified tools:

rules4 generate --lang go --tags "code style"

This command will:

  • Read the list of tools from your .rules4rc file.
  • Generate rules for each tool, focusing on code style for Go projects.

Command-Line Options

  • --primary <model_name>: Specify the primary AI model for rule generation (e.g., gpt-4-turbo, gpt4.1).
  • --review <model_name>: Specify an AI model for reviewing and refining the generated rules (e.g., claude-4-sonnet).
  • --research: Enable research using Perplexity AI before rule generation.
  • --lang <language>: Specify the programming language for rule generation (e.g., python, javascript, go).
  • --tags <tag1,tag2,...>: Comma-separated list of tags or topics for rule generation (e.g., pytest,langgraph, react,typescript, code style).
  • --dry-run: Preview the changes without actually writing any files.
  • --yes, -y: Overwrite existing files without prompting for confirmation.
  • --project-path <path>: (Optional) Specify the target project directory. Defaults to the current directory.

This project is in early development. For contributions, see CONTRIBUTING.md.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

rules4-0.0.5.tar.gz (11.5 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

rules4-0.0.5-py3-none-any.whl (8.3 kB view details)

Uploaded Python 3

File details

Details for the file rules4-0.0.5.tar.gz.

File metadata

  • Download URL: rules4-0.0.5.tar.gz
  • Upload date:
  • Size: 11.5 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.12.11

File hashes

Hashes for rules4-0.0.5.tar.gz
Algorithm Hash digest
SHA256 77868118c156d656cec15a386f7053c75b1242c9c25853fd31fbf736890f0cd3
MD5 e9d2a8316af5c35e10bda48a3c7a983f
BLAKE2b-256 a3dbffaedc7816d916e4dfee06c126f3a4028995e8d01dde83d8914321c7d6c7

See more details on using hashes here.

File details

Details for the file rules4-0.0.5-py3-none-any.whl.

File metadata

  • Download URL: rules4-0.0.5-py3-none-any.whl
  • Upload date:
  • Size: 8.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.12.11

File hashes

Hashes for rules4-0.0.5-py3-none-any.whl
Algorithm Hash digest
SHA256 9481321a291435205feda8e2b5264d5278f87991d1dabc28f598fb45f1a8fd10
MD5 e4adbfa09153b416907f88085f4bddc0
BLAKE2b-256 2d941e7883c835973cd18ef3bcf0e5226990aed1cda30b94bc8527a29f7b7b4f

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page