← Latest products

Cheddar-bench

unsupervised benchmark for coding agents. Unsupervised benchmark for evaluating CLI coding agents on bug detection. I built a small benchmark to test CLI coding agents on blind bug detection.

CLIOpen source

8 stars · GitHubexternal site

Visit github.comexternal site

Reactions (1)

Similar products

by description similarity

  1. RLM-based local debugger for AI agent traces

    github.com

    Hierarchal Agent Loop Optimizer

    Hacker News27 pts

  2. agenteval

    github.com

    static analysis for AI coding instruction file

    Hacker News7 pts

  3. MuscleMemory

    github.com

    procedural memory for coding agents, proven by tests

    Hacker News7 pts

  4. Capn-hook for coding agents

    github.com

    don't grep the same mystery twice

    Hacker News29 pts

  5. IHMT

    github.com

    Infinite Hierarchical Memory Tree for AI Agents

    Hacker News8 pts