Open memory & reliability layer for AI agents + independent judging/verification (NACRE judge, caliber-bench meta-benchmark, preregistered criteria) — by Nautilus Platform
Latest release v3.3.0 — v3.3.0 — 判分机构产品面(assay 子包) · 4 Oct 2026
2 files for Codex, OpenCode and Claude Code: nautilus-compass AGENTS.md, nautilus-compass CLAUDE.md — 940 tokens loaded in every session.
AGENTS.md A 153 tok CLAUDE.md A 787 tok These files are chunxiaoxx/nautilus-compass's own configuration — they tell Codex, OpenCode and Claude Code how to work on this repository, so they are not mods to install elsewhere. Copy one as a starting point and replace the parts that are about this project.