ResearchClawBench: Evaluating AI Agents for Automated Research from Re-Discovery to New-Discovery

Autoresearch with PhD-level workflows and modular agent skills. Built for the autonomous AI Scientist.

AutoR takes a research goal, runs a fixed 8-stage pipeline with Claude Code, and requires explicit human approval after every stage before the workflow can continue.

karpathy auto research ported to mlx so you can use your Mac

Lite research agents on proposal, experiment and review

Enhance academic workflows by auditing papers, verifying citations, and analyzing experiments with a research integrity plugin for Claude Code.

LISFLOOD-FP implementation of model-agnostic self-calibrating loop. Inspired by AutoResearch

General-purpose autonomous research framework for AI agents. Inspired by Andrej Karpathy's autoresearch.

Run your own research lab that never sleeps

Optimize AI agents autonomously by iterating code changes and evals to improve performance using LangSmith observability and automated experiments.

Measure AI agents’ performance with standardized tests across 314 tasks, 33 domains, and 4 difficulty levels for clear, reproducible comparison.