Skip to main content

SLIM Tools

This section provides an overview of tools created by the SLIM community to help implement best practices in your projects.

SLIM Leaderboard​

The SLIM Leaderboard is a tool that generates scan reports of SLIM best practices compliance across GitHub repositories. It helps you measure, rank, and showcase how well a set of repositories follows SLIM best practices.

Features​

  • Generates compliance reports for GitHub repositories based on the SLIM Getting Started Checklist
  • Ranks repositories from most to least compliant with SLIM best practices
  • Supports configuration via JSON to specify individual repositories or scan entire organizations
  • Works with both GitHub.com and GitHub Enterprise repositories
  • Utilizes GraphQL and parallelized queries for optimized performance
  • Provides multiple output formats including tree, table, and markdown
  • Offers a verbose mode for additional statistical details and explanations

Getting Started​

Requirements​

  • Python 3.7+
  • GitHub personal access token (classic) with all permissions under the "repo" group, including security_events

Installation​

# Install via pip
pip install slim-leaderboard

# Verify installation
slim-leaderboard --version
slim-leaderboard --help

# Upgrade to latest version
pip install --upgrade slim-leaderboard

Configuration​

Create a configuration file (e.g., slim-config.json) to specify the repositories or organizations to scan:

{
  "targets": [
      {
        "type": "repository",
        "name": "https://github.com/nasa-ammos/slim"
      }
  ]
}

Basic Usage​

# Generate a report using default settings (tree format)
slim-leaderboard slim-config.json

# Generate a report using default settings (tree format) for an adhoc repository
slim-leaderboard --repositories https://github.com/riverma/terraformly

# Generate a report in table format
slim-leaderboard --output_format TABLE slim-config.json

# Generate a verbose report in tree format
slim-leaderboard --output_format TREE --verbose slim-config.json

# Generate a markdown report with emojis instead of text for statuses
slim-leaderboard --output_format MARKDOWN --emoji slim-config.json

Demo​

Watch SLIM Leaderboard in action:

This demo demonstrates the key features of SLIM Leaderboard:

  • Version check: Shows the current version of slim-leaderboard
  • Help display: Shows available command-line options and usage
  • Individual repository scanning: Scans a single repository (Terraformly) using the --repositories argument with tree output format and emoji status indicators
  • Config file creation: Creates a JSON configuration file for scanning multiple NASA-AMMOS repositories
  • Batch scanning: Scans multiple repositories (SLIM, SLIM Leaderboard, and SLIM Starterkit Python) using the configuration file approach

🔄 Restart Demo

SLIM Leaderboard Demo - Scanning Repository

Learn More​

For more detailed information about SLIM Leaderboard, visit the GitHub repository.

SLIM CLI​

SLIM CLI is a command-line tool designed to infuse SLIM best practices seamlessly into your development workflow. It fetches and applies structured SLIM best practices directly into your Git repositories and leverages artificial intelligence capabilities to customize and tailor the application of best practices based on your repository's specifics.

⚠️ Important Changes in v2.0.0​

Legacy AI support has been deprecated in favor of the MCP (Model Context Protocol) plugin approach:

  • ❌ Deprecated: Direct --use-ai <model> with LiteLLM integration (will show warnings)
  • ✅ New Approach: Use --use-ai mcp to generate coordination prompts for MCP-enabled coding agents
  • 🔧 Migration: Install the MCP plugin to use AI features with Claude Code, Cursor, Windsurf, Aider, or other compatible coding agents
# Old approach (deprecated, will show warnings):
slim apply --best-practice-ids readme --repo-urls <repo> --use-ai ollama/llama3.1

# New approach (recommended):
slim apply --best-practice-ids readme --repo-urls <repo> --use-ai mcp
# Then copy the generated prompts to your coding agent

For seamless AI integration, use the MCP plugin with Claude Code, Cursor, Windsurf, Aider, or other compatible coding agents.

Demo​

Watch SLIM CLI in action:

This demo showcases the core functionality of SLIM CLI:

  • Version check: Displays the current version of SLIM CLI
  • Best practices listing: Shows all available SLIM best practices that can be applied
  • Local repository setup: Creates a sample repository and initializes it with Git
  • Local best practice application: Applies the README best practice to a local directory using --repo-dir
  • File inspection: Shows the generated README.md file contents
  • Remote repository application: Applies the README best practice to a GitHub repository (Terraformly) using --repo-urls

🔄 Restart Demo

SLIM CLI Demo - Apply Command

Features​

  • Modern CLI Interface: List, patch, and infuse SLIM best practices into your Git repository workflow using a seamless terminal interface
  • Fetches the latest SLIM best practices dynamically from SLIM's registry the moment they change
  • Patches and pushes, SLIM best practices to your repository and pushes up to your Git remote (i.e. GitHub) - all automatically
  • AI Enabled: Integration with coding agents through MCP plugin for AI-powered best practice customization
  • Extensible Architecture: Easy-to-extend best practice system with centralized mapping and YAML configuration

Getting Started​

Requirements​

  • Python 3.10+ (recommended for MCP server integration)
  • Git
  • Optional: For AI features, install the MCP plugin to work with coding agents (Claude Code, Cursor, Windsurf, Aider)

Installation​

# Install via pip (recommended for most users)
pip install slim-cli

# Verify installation
slim --version
slim --help

Basic Usage​

# List all available best practices
slim list

# Apply best practices to repositories (using aliases)
slim apply --best-practice-ids readme --best-practice-ids governance-small --repo-urls https://github.com/your-username/your-repo

# Generate AI coordination prompts for coding agents
slim apply --best-practice-ids readme --repo-urls https://github.com/your-username/your-repo --use-ai mcp

# Apply best practices and then push those changes to a new repository branch on GitHub.com - all automatically
slim apply-deploy --best-practice-ids readme --best-practice-ids governance-small --repo-urls https://github.com/your-username/your-repo --remote origin --commit-message "Apply SLIM best practices"

# Explore available AI models (for reference)
slim models list
slim models setup anthropic

Documentation Generation​

The SLIM CLI includes a website generator that can automatically create Docusaurus documentation from your repository content.

Demo​

🚧 Coming Soon - Updated demo video featuring the new MCP plugin workflow will be available in a future release.

Example Usage​

# Generate documentation with AI coordination prompts for coding agents
slim apply --best-practice-ids docs-website --repo-dir /path/to/your/repo --output-dir /path/to/output --use-ai mcp

Recommended Approach:

  • Use --use-ai mcp to generate coordination prompts
  • Copy the prompts to your coding agent (Claude Code, Cursor, Windsurf, Aider)
  • Let your coding agent handle the AI-powered customization

Example MCP Prompt: "Apply the SLIM docs-website best practice to generate a Docusaurus documentation site for this repository. Customize the content based on the project structure and README."

Unit Test Generation​

🚧 Coming Soon - The unit test generation feature is currently being refactored and will be available in a future release.

AI Models Discovery​

SLIM CLI supports model discovery and recommendations for use with coding agents:

# List all available models (100+ models from various providers)
slim models list

# Filter by provider
slim models list --provider anthropic
slim models list --provider openai

# Get AI model recommendations by task and quality tier
slim models recommend                              # Default: documentation, balanced
slim models recommend --task documentation --tier premium
slim models recommend --task code_generation --tier fast

# Get setup instructions for specific providers
slim models setup anthropic
slim models setup openai
slim models setup ollama

# Validate model configuration and test connectivity
slim models validate anthropic/claude-3-5-sonnet-20241022
slim models validate openai/gpt-4o
slim models validate ollama/llama3.1

Supported AI Providers:

  • Cloud Premium: OpenAI, Anthropic Claude, Google Gemini
  • Cloud Fast: Groq, Together AI, Cohere, Perplexity
  • Local/Private: Ollama, VLLM, LM Studio, GPT4All
  • Enterprise: Azure OpenAI, AWS Bedrock, Google Vertex AI

MCP Plugin Integration​

SLIM CLI includes a Model Context Protocol (MCP) plugin that enables seamless integration with coding agents like Claude Code, Cursor, Windsurf, Aider, and other MCP-compatible tools.

Key Features​

  • Natural Language Interface: Apply SLIM best practices using conversational AI commands
  • Repository Context Awareness: Automatically analyzes your repository structure and customizes templates
  • Real-time Best Practices: Fetches the latest SLIM best practices from the live registry
  • AI-Powered Customization: Leverages your coding agent's capabilities for intelligent template customization

Quick Setup for Claude Code​

  1. Install SLIM CLI (if not already installed):

    pip install slim-cli

  2. Add the MCP server to Claude Code - see the GitHub repository for detailed setup instructions

  3. Use in your coding agent:

    • Simply ask: "Apply the SLIM readme best practice with AI customization"
    • Or: "List all available SLIM best practices"
    • Or: "Apply governance template to this repository"

Learn More​

For more detailed information about SLIM CLI, including MCP setup and migration guides, visit the GitHub repository.