terminal-bench

What problem does it solve?

This Skill provides strategies and guidance for successfully completing the diverse range of tasks found in the Terminal-Bench 2.0 benchmark, ensuring efficient and accurate execution.

Core Features & Use Cases

• Task Classification: Automatically categorizes tasks based on keywords (e.g., git, compression, chess) to apply relevant strategies.
• Universal Principles: Offers a set of best practices for coding and problem-solving applicable across all tasks.
• Pre-Execution Hooks: Guides on backing up critical system states before potentially disruptive operations.
• Use Case: When faced with a complex git recovery task, this Skill will help classify it, remind you of universal principles like reading the full task first, and suggest backing up your .git directory before proceeding.

Quick Start

Use the terminal-bench skill to classify the task 'git fsck' and apply relevant strategies.