minimal-run-and-audit: What Its Scripts Actually Record
We ran RigorPilot’s minimal-run-and-audit scripts on a real repo: the metric it calls best, the shell-mode gate, and a pipe that fakes success.
RigorPilot’s 49-line helper hit 450,765 installs by refusing most paper requests: one gap per answer, README conflicts recorded, never overridden.
We ran RigorPilot’s minimal-run-and-audit scripts on a real repo: the metric it calls best, the shell-mode gate, and a pipe that fakes success.
RigorPilot’s repo-intake-and-plan (450K installs, #70) scanned micrograd, makemore, and nanoGPT for me: 2, 3, and 16 commands — plus one it invented.
RigorPilot’s ai-research-reproduction (310.8K installs) runs DL repos as 9 README-first steps: smallest target, audited patches, repro_outputs/ evidence.