awesome-autoresearch and autoresearch
These are ecosystem siblings where A is a curated discovery/reference resource for the autoresearch paradigm while B is a specific implementation of that paradigm for Claude Code plugins, allowing practitioners to find and apply autoresearch concepts.
About awesome-autoresearch
alvinunreal/awesome-autoresearch
A curated list of autonomous improvement loops, research agents, and autoresearch-style systems inspired by Karpathy's autoresearch.
Organizes 50+ implementations across general-purpose descendants, research-agent systems, domain-specific adaptations, and hardware ports—each with distinct architectural patterns like closed-loop evaluation, swarm coordination, prompt evolution, and tree-search optimization. Covers both narrow autoresearch forks (Gemini-native with live search grounding, Claude Code skills with resumable sessions) and broader meta-systems that auto-generate agent architectures or orchestrate multi-loop optimization. Includes recent ICLR papers (ADAS, GEPA, ClawTeam) demonstrating self-improvement on coding benchmarks and prompt tuning via genetic-Pareto reflection.
About autoresearch
zircote/autoresearch
Autonomous skill improvement loop for Claude Code plugins — inspired by Karpathy's autoresearch. Modify → evaluate → keep/discard → repeat until convergence. Zero-touch quality iteration at scale.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work