DAFC (The Fuck?) CLI - Dumb as Fuck Coding Tool

DAFC is a command-line tool and a methodology designed to leverage the massive context windows of modern Large Language Models (LLMs) like Gemini 1.5 Pro. Instead of complex context management, DAFC lets you easily dump your entire (but small!) codebase into the LLM prompt for querying, code generation, analysis, and more.

Read the full philosophy and background in the introductory blog post.

Core Idea: Constraints Unlock Full Context

The "Dumb as Fuck Coding" methodology hinges on keeping projects simple and constrained:

• Max ~500 Lines Per File (Aim for ~300): Keeps individual units manageable.
• Max ~50 Files (Aim for ~10): Limits overall scope.
• Max ~5 Database Tables (Aim for 3): Simplifies the data model. NOTE: Does not include users table.

By adhering to these rules, the entire relevant codebase often fits within the LLM's context window. DAFC CLI automates gathering this context, checks if it exceeds a configured limit, and interacts with the LLM.

Features

• Simple CLI Interface: Easy-to-use commands (ask, context, config, init).
• Automatic Context Gathering: Recursively scans your project, respects .gitignore and .dafcignore, and includes relevant files.
• Context Limit Check: Estimates token count and throws an error if it exceeds the configured maximum (MAX_CONTEXT_TOKENS).
• Layered Configuration: Uses global (~/.config/dafc/config.env), project (./.env), and environment variables for flexible settings (Env > Project > Global).
• Easy Config Editing: dafc config command to open config files in your default editor.
• Customizable Rules: Uses a .dafcr file to inject system prompts or specific instructions into the LLM request.
• Flexible LLM Backend: Configurable via config files/env vars to use any OpenAI-compatible API (OpenRouter, Featherless, OpenAI, Google AI Studio, etc.). Defaults to OpenRouter with google/gemini-1.5-pro-latest.
• LLM Interaction: Sends context + prompt to the configured LLM. Includes streaming output and retries.
• Response Handling: Streams the LLM's response to your console and saves the full response to response.md.
• Easy Installation: Shell script, package managers (npm, yarn, bun).

Installation

Using the Install Script (Linux/macOS)

1. Make sure you have Bun (install instructions) and Git installed.

2. Run the following command in your terminal. Installs to /usr/local/bin by default:

   curl -fsSL https://raw.githubusercontent.com/AviSantoso/dafc/main/install.sh | sudo bash

   • NOTE: Uses sudo for default install location. Review the script if needed.
   • Alternatively, modify the INSTALL_DIR variable in the script before running, or install via package manager.

3. Follow any on-screen instructions, especially regarding adding the installation directory to your PATH if needed.

4. Restart your terminal or source your shell profile (source ~/.bashrc, source ~/.zshrc, etc.).

5. Verify installation: dafc --version

Using NPM/Yarn/Bun Package Managers

1. Install globally with your preferred package manager:

   # Using npm
   npm install -g @wagn/dafc
   
   # Using yarn
   yarn global add @wagn/dafc
   
   # Using bun
   bun install -g @wagn/dafc

2. Verify installation:

   dafc --version

Manual Installation / From Source

1. Ensure Bun and Git are installed.
2. Clone the repository:

   git clone https://github.com/AviSantoso/dafc
   cd dafc

3. Install dependencies:

   bun install --frozen-lockfile

4. Build the executable:

   bun run build

5. Move the generated dafc executable to a directory in your PATH (e.g., /usr/local/bin, ~/.local/bin, ~/.bun/bin).

   # Example:
   mv ./dafc ~/.local/bin/
   # Ensure ~/.local/bin is in your PATH

Configuration

DAFC uses a layered configuration system:

1. Environment Variables (Highest Priority): Any setting prefixed with OPENAI_ or other config keys (e.g., TEMPERATURE, MAX_CONTEXT_TOKENS) set in your shell environment will override all other settings.

   export OPENAI_API_KEY='your-key-here'
   export OPENAI_MODEL='anthropic/claude-3.5-sonnet'

2. Project .env File (Medium Priority): Create a .env file in your project's root directory. Settings here override the global config. This is the recommended place for project-specific LLM choices or API keys.

   # .env (in your project root)
   OPENAI_API_KEY='project-specific-key-maybe'
   OPENAI_MODEL='openai/gpt-4o'

   • Use dafc config to easily create or edit this file.

3. Global Config File (Lowest Priority): A global file stores default settings for all projects. Located at:

   • Linux/macOS: ~/.config/dafc/config.env
   • Windows: %APPDATA%\dafc\Config\config.env (e.g., C:\Users\YourUser\AppData\Roaming\dafc\Config\config.env)

   # ~/.config/dafc/config.env (Example)
   OPENAI_API_KEY='your-default-openrouter-key'
   OPENAI_MODEL='google/gemini-1.5-pro-latest'
   OPENAI_API_BASE='https://openrouter.ai/api/v1'

   • Use dafc config --global to easily create or edit this file.

Key Configuration Variables:

• OPENAI_API_KEY (Required): Your API key for the chosen service.
• OPENAI_MODEL: The specific LLM model identifier (e.g., google/gemini-1.5-pro-latest, openai/gpt-4o, anthropic/claude-3.5-sonnet).
• OPENAI_API_BASE: The base URL for the LLM API endpoint (e.g., https://openrouter.ai/api/v1, https://api.openai.com/v1, https://api.featherless.ai/v1).
• TEMPERATURE: LLM creativity/randomness (0.0-2.0).
• MAX_CONTEXT_TOKENS: Safety limit for context size (default ~900k).
• MAX_FILE_SIZE_BYTES: Skip files larger than this (default 1MB).

Example Setups:

• OpenRouter (Default): Set OPENAI_API_KEY in global or project config. OPENAI_API_BASE defaults to OpenRouter. Choose a model via OPENAI_MODEL.
• OpenAI: Set OPENAI_API_KEY, OPENAI_MODEL (e.g., gpt-4o), and OPENAI_API_BASE='https://api.openai.com/v1'.
• Featherless: Set OPENAI_API_KEY, OPENAI_MODEL (e.g., openai/deepseek-ai/DeepSeek-V3-0324), and OPENAI_API_BASE='https://api.featherless.ai/v1'.
• Google AI Studio: Set OPENAI_API_KEY, OPENAI_MODEL (e.g., gemini-1.5-pro-latest), and OPENAI_API_BASE='https://generativelanguage.googleapis.com/v1beta'. Note: Google's API might require specific formatting or headers not fully tested.

Initialize Helper Files:

Run dafc init in your project's root directory. This creates:

• .dafcignore: Add file/directory patterns (like .gitignore) to exclude from the context sent to the LLM.
• .dafcr: Define custom system prompts or rules for the LLM.

Usage

Run dafc commands from your project's root directory.

1. Ask the LLM (dafc ask <prompt> or dafc <prompt>)

The primary command. Gathers context, checks limits, includes rules from .dafcr, sends to LLM, streams response.

# Ask a question (default command is 'ask')
dafc "How does the authentication logic in user.ts work?"

# Request code generation
dafc ask "Generate a new API endpoint in routes/items.ts to delete an item by ID."

# Ask for refactoring help
dafc a "Refactor the main function in cli.ts to be more modular." # Using alias

# Get help running the project
dafc "What are the steps to run this project locally?"

Response streams to terminal, saved to response.md. Aborts if MAX_CONTEXT_TOKENS exceeded. Use --debug to see resolved config.

2. Manage Context (dafc context)

View, save, copy, or watch the gathered context. Respects MAX_CONTEXT_TOKENS.

# Print context to console (default)
dafc context

# Save context to context.md
dafc context --save

# Save context to a specific file
dafc context -s my_context.txt # Using alias

# Copy context to clipboard
dafc context --copy

# Save to context.md AND copy
dafc context --save --copy

# Save to context.md and watch for file changes, updating automatically
# Press Ctrl+C to stop.
dafc context --save --watch

# Save to specific file and watch
dafc context -s specific_context.txt -w

• -s, --save [filename]: Saves context. Defaults to context.md if no filename given.
• --copy: Copies context to clipboard.
• -w, --watch: Requires --save. Monitors project and updates saved file on changes.

3. Edit Configuration (dafc config)

Opens the relevant configuration file in your default text editor.

# Edit the project-specific .env file (creates if needed)
dafc config

# Edit the global config file (creates if needed)
dafc config --global
dafc config -g # Using alias

4. Initialize Helper Files (dafc init)

Creates .dafcignore and .dafcr with default templates if they don't exist.

dafc init

How It Works

dafc ask "prompt"

1. Loads config (Env > Project .env > Global config.env).
2. Reads .gitignore and .dafcignore.
3. Scans current directory recursively for allowed files.
4. Filters out ignored/empty/oversized files.
5. Estimates total token count (files + boilerplate).
6. Aborts if MAX_CONTEXT_TOKENS exceeded.
7. Reads rules from .dafcr (if exists).
8. Formats context block (file contents, paths, sizes, etc.).
9. Sends [System Prompt from .dafcr (if any)] + [User Message (Context Block + User Prompt)] to the configured LLM API.
10. Streams response to stdout and saves full response to response.md.

Contributing

Contributions welcome! Please submit pull requests or open issues.

Disclaimer

This tool sends your code (excluding ignored files) to external LLMs.

• Do not use with sensitive codebases. Keep secrets out of code, use ignore files, and manage API keys securely.
• Use at your own risk. Always review LLM-generated code carefully.

License

MIT License - see LICENSE.