wut is a terminal assistant that leverages large language models (LLMs) to explain the output of previous shell commands, aiding users in understanding errors, deciphering logs, and fixing incorrect commands. It targets developers and power users seeking to quickly interpret complex terminal outputs.

How It Works

The CLI tool captures the output of the last executed command within a tmux or screen session. It then sends this output, along with an optional user query, to a configured LLM provider (OpenAI, Anthropic, or Ollama) for analysis and explanation. This approach provides on-demand context-aware assistance directly within the terminal workflow.

Quick Start & Requirements

• Install via pipx install wut-cli.
• Requires tmux or screen session to capture command output.
• Supports OpenAI, Anthropic (Claude), and Ollama (local models) via API keys or environment variables (OPENAI_API_KEY, ANTHROPIC_API_KEY, OLLAMA_MODEL).
• Optional OpenAI configuration via OPENAI_MODEL and OPENAI_BASE_URL.

Highlighted Details

• Explains stack traces, error codes, and incorrect commands.
• Supports custom queries for specific command-related questions.
• Integrates with popular LLM providers and local models.

Maintenance & Community

• Project is maintained by shobrook.
• Roadmap includes removing tmux/screen dependency, adding an auto-fix option, and Homebrew integration.

Licensing & Compatibility

• The provided README does not specify a license.

Limitations & Caveats

• Currently requires running within a tmux or screen session to function.
• The project is in active development with a roadmap indicating planned features and improvements.