autoresearch vs loopy
Karpathy's autoresearch loop as an installable skill for Claude Code, OpenCode and Codex: constraint + mechanical metric + autonomous modify→verify→keep/discard iteration. — versus — A public library of reusable AI-agent loops plus Loopy, an installable skill that helps agents find, audit, adapt, run and publish loops from the live catalog.
Both distribute reusable agent loops: autoresearch is one loop (Karpathy's research iteration) as an installable skill; Loopy is a whole catalog of loops plus find/audit/run tooling.
| autoresearch | loopy | |
|---|---|---|
| Stars | 5.3k | 2.7k |
| Forks | 395 | 235 |
| Language | Shell | JavaScript |
| License | MIT | MIT |
| Last activity | 23 days ago | 9 days ago |
| Topics | skills, coding | skills |
| Curated connections | 2 | 2 |
autoresearch — the curator's take
The compounding-gains loop, packaged: pick a metric a machine can check, let the agent mutate-verify-keep against it for hours, and small wins stack — the Karpathy recipe without writing the harness yourself. Works across three agent CLIs. The discipline it demands is the catch: without a truly mechanical metric the loop optimizes noise, and unattended iteration burns real tokens — set budgets before you set it loose.
loopy — the curator's take
The insight: most agent work is a repeatable loop someone already designed — so catalog them. The website is browsable by humans AND agents (llms.txt, JSON catalog, agent guide), and the Loopy skill turns your agent into a loop librarian: discover, audit, repair, debrief, publish. Genuinely useful for not reinventing the same research/review/refactor loop weekly. NOT a runtime — loops are prompts and procedure, not executable infrastructure; quality varies by contributor, so audit before you adopt (the skill's audit step exists for a reason).