← ClaudeAtlas

ship-reviewlisted

Unified pre-PR review — multi-model code review (gstack Claude + Codex everywhere; WORK adds pal Gemini + GPT-5, PERSONAL adds a pal-free personal-key Gemini pass; work/personal aware) plus a user-journey pass that walks the whole flow through in order as a first-time user, so a diff that reads fine but leaves a dead end, a missing empty/loading/error state, or a step nobody can reach is caught before it ships. Use for "code review", "review my changes", "walk the whole flow", "walk through this as a user", "pre-ship UX pass". Judges work that EXISTS; whether it should be built at all is /sk:work-does-this-make-sense-to-build.
skowalskidev/dotclaude · ★ 2 · Code & Development · score 78
Install: claude install-skill skowalskidev/dotclaude
Run Simon's unified pre-PR review: combine the gstack review with the pal multi-model review, walk the change as a real user, and STRICTLY respect work/personal resource isolation. **Running the journey pass alone.** Steps 1–5 are the multi-model code review; Step 6 is the user-journey pass. When the ask is only "walk this as a user" (no code review wanted), run Step 6 on its own and say that is what you did — don't spend the model passes to get there. ## Step 1 — Detect context (required) - Repo origin: !`git remote get-url origin 2>/dev/null || echo "(no git remote)"` - Diff scope: !`base=$(git merge-base origin/master HEAD 2>/dev/null || git merge-base origin/main HEAD 2>/dev/null || echo HEAD~1); git diff --stat "$base"...HEAD 2>/dev/null | tail -1` Rule: if the origin matches your work org (workOrgMatch in your identity overlay) → **WORK** repo. Otherwise → **PERSONAL** repo. Before reviewing, state in one line the detected context and exactly which reviewers/resources you will use. ## Step 2 — Run the reviewers for that context **Cover the WHOLE diff — every changed file, not a risk-weighted subset.** Feed the reviewers the entire diff from Step 1, tests and UI and docs included. When it is too large for one model call, PARTITION it so every file is covered by some reviewer (fan out across parallel passes — `/sk:work-superspeed` slices each owning a share of the files is one clean cut), never sample down to the "high-risk" files and read the rest by hand or not at