rubber-ducky 1.5.3

Quick CLI do-it-all tool. Use natural language to spit out bash commands

Rubber Ducky

Rubber Ducky is an inline terminal companion that turns natural language prompts into runnable shell commands. Paste multi-line context, get a suggested command, and run it without leaving your terminal.

Quick Start

Action	Command
Install globally	uv tool install rubber-ducky
Run once	uvx rubber-ducky -- --help
Local install	uv pip install rubber-ducky

Requirements:

• Ollama running locally or use cloud models
• Model available via Ollama (default: glm-4.7:cloud)

Usage

ducky                      # interactive inline session
ducky --directory src      # preload code from a directory
ducky --model qwen3        # use a different Ollama model
ducky --local              # use local models with qwen3 default

Both ducky and rubber-ducky executables map to the same CLI, so uvx rubber-ducky -- <args> works as well.

Inline Session (default)

Launching ducky with no arguments opens the inline interface:

• Enter submits; Ctrl+J inserts a newline (helpful when crafting multi-line prompts). Hitting Enter on an empty prompt reruns the latest suggested command; if none exists yet, it explains the most recent shell output.
• Ctrl+R re-runs the last suggested command.
• Ctrl+S copies the last suggested command to clipboard.
• Prefix any line with ! (e.g., !ls -la) to run a shell command immediately.
• Arrow keys browse prompt history, backed by ~/.ducky/prompt_history.
• Every prompt, assistant response, and executed command is logged to ~/.ducky/conversation.log.
• Press Ctrl+D on an empty line to exit.
• Non-interactive runs such as cat prompt.txt | ducky print one response (and suggested command) before exiting; if a TTY is available you'll be asked whether to run the suggested command immediately.
• If prompt_toolkit is unavailable in your environment, Rubber Ducky falls back to a basic input loop (no history or shortcuts); install prompt-toolkit>=3.0.48 to unlock the richer UI.

ducky --directory <path> streams the contents of the provided directory to the assistant the next time you submit a prompt (the directory is read once at startup).

Model Management

Rubber Ducky now supports easy switching between local and cloud models:

• /model - Interactive model selection between local and cloud models
• /local - List and select from local models (localhost:11434)
• /cloud - List and select from cloud models (ollama.com)
• Last used model is automatically saved and loaded on startup
• Type esc during model selection to cancel

Additional Commands

• /help - Show all available commands and shortcuts
• /crumbs - List all saved crumb shortcuts
• /crumb <name> - Save the last AI-suggested command as a named crumb
• /crumb add <name> <command> - Manually add a crumb with a specific command
• /crumb del <name> - Delete a saved crumb
• <crumb-name> - Invoke a saved crumb (displays info and executes the command)
• /clear or /reset - Clear conversation history
• /run or :run - Re-run the last suggested command

Crumbs

Crumbs are saved command shortcuts that let you quickly reuse AI-generated bash commands without regenerating them each time. Perfect for frequently-used workflows or complex commands.

Saving Crumbs

When the AI suggests a command that you want to reuse:

1. Get a command suggestion from ducky
2. Save it immediately: /crumb <name>
3. Example:

   >> How do I list all Ollama processes?
   ...
   Suggested command: ps aux | grep -i ollama | grep -v grep
   >> /crumb ols
   Saved crumb 'ols'!
   Generating explanation...
   Explanation added: Finds and lists all running Ollama processes.
   

The crumb is saved with:

• The original command
• An AI-generated one-line explanation
• A timestamp

Invoking Crumbs

Simply type the crumb name in the REPL or use it as a CLI argument:

In REPL:

>> ols

Crumb: ols
Explanation: Finds and lists all running Ollama processes.
Command: ps aux | grep -i ollama | grep -v grep

$ ps aux | grep -i ollama | grep -v grep
user123  12345  0.3  1.2  456789  98765 ?  Sl  10:00   0:05 ollama serve

From CLI:

ducky ols              # Runs the saved crumb and displays output

When you invoke a crumb:

1. It displays the crumb name, explanation, and command
2. Automatically executes the command
3. Shows the output

Managing Crumbs

List all crumbs:

>> /crumbs

Output:

Saved Crumbs
=============
ols      | Finds and lists all running Ollama processes. | ps aux | grep -i ollama | grep -v grep
test     | Run tests and build project                  | pytest && python build.py
deploy   | Deploy to production                         | docker push app:latest

Manually add a crumb:

>> /crumb add deploy-prod docker build -t app:latest && docker push app:latest

Delete a crumb:

>> /crumb ols
Deleted crumb 'ols'.

Storage

Crumbs are stored in ~/.ducky/crumbs.json as JSON. Each crumb includes:

• prompt: Original user prompt
• response: AI's full response
• command: The suggested bash command
• explanation: AI-generated one-line summary
• created_at: ISO timestamp

Example:

{
  "ols": {
    "prompt": "How do I list all Ollama processes?",
    "response": "To list all running Ollama processes...",
    "command": "ps aux | grep -i ollama | grep -v grep",
    "explanation": "Finds and lists all running Ollama processes.",
    "created_at": "2024-01-05T10:30:00.000000+00:00"
  }
}

Delete ~/.ducky/crumbs.json to clear all saved crumbs.

Development (uv)

uv sync
uv run ducky --help

uv sync creates a virtual environment and installs dependencies defined in pyproject.toml / uv.lock.

Telemetry & Storage

Rubber Ducky stores:

• ~/.ducky/prompt_history: readline-compatible history file.
• ~/.ducky/conversation.log: JSON lines with timestamps for prompts, assistant messages, and shell executions.
• ~/.ducky/config: User preferences including last selected model.

No other telemetry is collected; delete the directory if you want a fresh slate.