LLM Loop — AI skill for Claude Code

LLM Loop

**An autonomous task execution plugin for the LLM CLI tool**

`llm loop` is a powerful plugin for the [LLM CLI tool](https://github.com/simonw/llm) that enables autonomous, goal-oriented task execution. Unlike traditional single-turn LLM interactions, `llm loop` allows the AI to work persistently towards a goal by making multiple tool calls, analyzing results, and iterating until the task is complete.


🚀 Features

• Goal-Oriented Execution: Define a task and let the AI work autonomously to complete it
• Tool Integration: Seamlessly use built-in tools and custom Python functions
• Iterative Problem Solving: Chain multiple tool calls and adapt based on results
• Interactive Control: Configure turn limits, approve tool calls, and guide the process
• Safety Features: Built-in approval mechanisms for potentially dangerous operations
• Conversation Export/Import: Save and reload conversation history

📦 Installation

Prerequisites

1. **Install LLM CLI globally** (recommended using `uv`):

   uv tool install llm

   Or using `pipx`:

   pipx install llm

2. **Configure an LLM model** (you'll need an API key):

   # For OpenAI
   llm keys set openai
   llm models default gpt-4.1-mini
   
   # For Anthropic
   llm keys set anthropic
   llm models default claude-3-5-sonnet-20241022

Install the Plugin

**Option 1: Install from PyPI (Recommended for most users)**

Once the plugin is available on PyPI, you can install it directly using `llm` or `pip`:

# Using llm (recommended)
llm install llm-loop-plugin

# Or using pip
pip install llm-loop-plugin

**Option 2: Install from source (for development or latest changes)**

1. **Clone this repository**:

   git clone https://github.com/nibzard/llm-loop
   cd llm-loop

2. **Install the plugin**:

   # For regular use from source
   llm install -e .