autoresearch-prompt

Community

Optimize prompts with AI-driven eval loops.

Authormaroffo
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill automates the process of improving AI prompts by creating an iterative evaluation and refinement loop, ensuring prompts are highly effective for specific tasks.

Core Features & Use Cases

  • Autonomous Prompt Optimization: Uses a Karpathy-style autoresearch pattern to continuously improve prompts based on evaluation results.
  • Eval-Driven Experimentation: Runs prompts against labeled examples, analyzes failures, and makes targeted edits to prompt.md.
  • Use Case: You have a prompt that isn't reliably classifying newsletter content. This Skill will automatically test variations of your prompt against a dataset of newsletters, identify why it fails, and iteratively refine it until it achieves high accuracy.

Quick Start

Run the autoresearch-prompt skill to evaluate the current prompt against the default evaluation set.

Dependency Matrix

Required Modules

anthropicpydantic

Components

scriptsreferencesassets

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: autoresearch-prompt
Download link: https://github.com/maroffo/claude-forge/archive/main.zip#autoresearch-prompt

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.