From 84a93766f21bbcff70579c5a818f10786910a07c Mon Sep 17 00:00:00 2001 From: Garry Tan Date: Sat, 15 Aug 2026 16:49:40 -0700 Subject: [PATCH] docs: CLAUDE.md describes the shipped shard-process model, not the abandoned --parallel probe --- CLAUDE.md | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/CLAUDE.md b/CLAUDE.md index 293862a8e..ac661a717 100644 --- a/CLAUDE.md +++ b/CLAUDE.md @@ -96,9 +96,10 @@ bun run test # run before every commit — free, ~90-100s for the full ~ bun run test:evals # run before shipping — paid, diff-based (~$4/run max) ``` -`bun run test` routes through `scripts/test-free-shards.ts` (one `bun test ---parallel` invocation with strict-output classification: a run without bun's -terminal summary line, or with a crashed worker, FAILS — silent truncation +`bun run test` routes through `scripts/test-free-shards.ts` (N concurrent +shard processes, serial within each, plus a trailing serial tree-mutating +shard — with strict-output classification per shard: a shard without bun's +terminal summary line FAILS — silent truncation cannot report green). Never type bare `bun test` for the suite: it walks the whole repo, loading paid eval files and missing the strict classifier. It covers skill validation, gen-skill-docs quality checks, and browse