[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7586J4ZNK203RJYV36GECG8":3,"skill-same:01M7586J4ZNK203RJYV36GECG8":46},{"skillNo":4,"name":5,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":18,"linkOnly":19,"markdownContent":20,"markdownBody":21,"frontMatter":22,"refName":23,"commitSha":24,"skillDirectory":25,"forksCount":26,"repositoryHtmlUrl":27,"repositoryHomepageUrl":28,"ownerHtmlUrl":29,"ownerBlogUrl":28,"ownerDescription":30,"updatedAt":31,"install":32,"resources":45},"01M7586J4ZNK203RJYV36GECG8","corpus-remeasure",null,"nubjs\u002Fnub","nubjs","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F6785695?v=4","01M7583N3EYYHSNB6FJQXCDMQH","Nub","01M7583N44MVMSN8385QEF3C8V",4440,"MIT","https:\u002F\u002Fgithub.com\u002Fnubjs\u002Fnub\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fcorpus-remeasure\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: corpus-remeasure\ndescription: The iteration cycle for the build-jail corpus — change the harness or nub, push it, re-measure ONE package@version on ONE platform, and read the answer. Invoke (via the Skill tool) whenever you need to test a hypothesis about why a package got the grant it did, verify a harness or nub fix on a real runner, or re-measure specific packages without disturbing the corpus. Carries the traps that each cost a wasted run: a bare `git commit` that clobbers the queue, `--force` argument order, a watcher that reports a pass off an empty log, and the fact that a grant is not evidence until you read the failing cell's log.\n---\n\n# Re-measuring a package in the build-jail corpus\n\nThe loop is: **edit → push → dispatch a debug run → watch → read the CELL LOGS**. Each step has a trap that has already cost a run.\n\nCorpus repo: `nubjs\u002Fbuild-jail-corpus` (public). Local clone: `~\u002Fnub-corpus-staging`. nub: branch `sandbox\u002Fintegration` at `~\u002F.cache\u002Fnub\u002Fworktrees\u002Finteg`.\n\n## 1. Push the harness change — the queue is the trap\n\n⛔ **ALWAYS `git commit -- \u003Cexplicit paths>`.** A bare `git commit` sweeps in whatever is staged, and `queue.ndjson` is almost always staged because reading the corpus stages it. A commit carrying a stale queue **releases rows other runners are actively measuring**. Measured: one such commit carried 194 lines of stale queue.\n\n```sh\ncd ~\u002Fnub-corpus-staging\ngit commit -q -m \"…\" -- harness\u002Fsearch.mjs          # explicit paths, always\ngit show --stat --oneline HEAD | head -5             # one second; catches exactly this\n```\n\nThe push races ~12 concurrent runners, so it needs a retry loop — and the cleanup before each attempt is what makes it work:\n\n```sh\nfor a in 1 2 3 4 5; do\n  git restore --staged . 2>\u002Fdev\u002Fnull                 # unstage (checkout STAGES; restore is what unstages)\n  git checkout -- records queue.ndjson 2>\u002Fdev\u002Fnull   # drop working-copy edits — origin owns these\n  git clean -fdq -- records 2>\u002Fdev\u002Fnull              # reading the corpus leaves UNTRACKED records behind\n  git fetch -q origin main\n  git rebase -q origin\u002Fmain >\u002Fdev\u002Fnull 2>&1 || { git rebase --abort 2>\u002Fdev\u002Fnull; sleep 3; continue; }\n  git push origin HEAD:main >\u002Fdev\u002Fnull 2>&1 && { echo \"PUSHED\"; break; }\n  sleep 3\ndone\ngit fetch -q origin main\ngit merge-base --is-ancestor HEAD origin\u002Fmain && echo \"✓ LANDED\" || echo \"⛔ NOT LANDED\"\n```\n\n⛔ **Verify it landed. Never infer it from the push command's apparent success** — read `merge-base --is-ancestor`, because a rejected push inside a loop is easy to miss.\n\n**Reading the corpus for analysis leaves two kinds of debris**, both of which break the next rebase: `git checkout origin\u002Fmain -- records` STAGES those files, and it leaves records your HEAD lacks as UNTRACKED. Before cleaning untracked records, confirm they exist at origin (`git cat-file -e origin\u002Fmain:\u003Cpath>`) so a clean can never destroy an unpushed measurement.\n\n## 2. Dispatch the re-measure\n\nThe workflow has a **`packages` debug input**. When set, the runner SKIPS the queue entirely — no claim, no completion, no commit of results — and measures exactly those specs with `--force`, publishing only the ARTIFACT. That is what makes it safe to run against a live corpus.\n\n```sh\nkeyward --reason \"\u003Cthe task>\" -- gh workflow run corpus-queue-runner.yml -R nubjs\u002Fbuild-jail-corpus \\\n  -f os=macos \\                       # linux | macos | windows\n  -f nub_sha=\u003C40-char sha> \\\n  -f chain=false \\                    # ⛔ or it self-dispatches a whole chain\n  -f slice=2 \\\n  -f packages=\"husky@3.1.0 detox@20.9.1\"\n```\n\n- ⛔ **`nub_sha` must be PUSHED.** CI clones `nubjs\u002Fnub` anonymously. Check with `git ls-remote https:\u002F\u002Fgithub.com\u002Fnubjs\u002Fnub sandbox\u002Fintegration` before dispatching, or the run dies at clone.\n- **Pick the sha deliberately.** Using the sha the fleet already built gets a binary-cache hit (~0 min instead of a ~26 min cold Windows build) AND makes the result comparable with corpus records. Use a NEW sha only when testing a nub change.\n- **Platform speed differs enormously.** macOS\u002FLinux ≈ 1 min\u002Fpackage; Windows ≈ 9 min\u002Fpackage average and up to ~40 for a package that walks the full 55-cell ladder, plus ~26 min for a cold build. **Prototype on macOS or Linux whenever the question is not Windows-specific.**\n- ⛔ **In `run-batch.sh`, `--force` must come BEFORE `--file`** — it is parsed only as the first argument after `\u003Cnub>`. The workflow already does this. A run with the order wrong \"succeeded\" in 17 seconds having measured nothing.\n\n## 3. Watch it — and gate the watcher, or it will lie to you\n\n⛔ **NEVER dispatch CI without a bounded background watcher.** A stall is silent. And the watcher itself must gate, because **`gh run view --log` returns an EMPTY string for an in-progress run** — so `grep -q \u003Cbad-thing>` returns false and reports a PASS off nothing. That exact bug appeared inside a gate written to prevent it.\n\nThree gates, in order, all mandatory:\n\n1. **Terminal state**, on a BOUNDED loop that exits non-zero on timeout (`for i in $(seq 1 240); do … sleep 45; done`) so it can never hang forever.\n2. **Log non-empty** — `wc -l` over 100 lines, else refuse to interpret.\n3. **The canary RAN** — assert `fixture canary OK: \u003Cn> files` is PRESENT, never infer it from the absence of `REFUSING TO RUN`.\n\nThe canary is the harness's own guard: it installs `puppeteer@25.4.0` and refuses the whole batch unless the fixture produces >5,000 files. **If it refuses, nothing below it means anything** — that is how a broken isolation change gets caught before it writes garbage.\n\n## 4. Read the CELL LOGS, not just the grant\n\n⛔ **THE SINGLE MOST EXPENSIVE LESSON HERE.** Three consecutive fixes for one Windows defect were wrong because each was theorised from a grant and a path list. The failing cell's log named the exact cause in one line, every time.\n\nPer-cell logs are gitignored, so they exist ONLY in the run artifact — and only when the batch actually ran:\n\n```sh\ngh run download \u003Crun-id> -R nubjs\u002Fbuild-jail-corpus -D \u002Ftmp\u002Fart -n records-\u003Cos>-\u003Crun-id>\nfind \u002Ftmp\u002Fart -path '*\u003Cpkg>*' -name '*.log'     # one log per cell, named for the state\n```\n\nRead the cell **one rung below the winner** — that is where the reason lives. Read it RAW before grepping: a grep that matches nothing looks identical to a clean run. Real examples, each of which redirected an investigation:\n\n| log line | what it actually meant |\n|---|---|\n| `spawnSync git EPERM` | needs to EXECUTE git — an exec denial, not a write denial |\n| `mkdir '…\\Packages\\nub_sbx_\u003Cpid>_\u003Cnonce>_0'` | the leaf is the PROFILE dir, not `Packages` — a fixture cannot pre-create it |\n| `Could not find JAVA, skipping …` | a real line that was the WRONG explanation — both platforms skipped identically |\n| *(no error at all)* | a silent partial install; the cell failed on DIGEST, not exit code |\n\n⛔ **A record's `grant` tells you WHAT was needed; only the log tells you WHY.** And a plausible log line can still be the wrong explanation — check whether it differs across the platforms that disagree.\n\n## 5. Attribute the result\n\n- **`corpusGitSha`** in `provenance` names the corpus commit exactly. Use it — `harnessSha256` is a content hash and git rewrites line endings on Windows, so the same commit hashes differently there and **cannot be compared across platforms**.\n- **Confounds before blaming a platform:** same `nubGitSha`? same `nodeSelection.chosenMajor`? And note darwin records are arm64 while linux records are x64, so OS and ARCH move together and cannot be separated from these records alone.\n- **A debug run writes NOTHING to the corpus** — results live only in the artifact. To make a re-measurement stick, let the fleet re-claim the row (reopen it) rather than hand-editing records.\n\n## Scope\n\n**Use it for** a hypothesis about one package, verifying a harness or nub fix on a real runner, or getting cell logs for a package whose artifact expired (30-day retention).\n\n**Not for** running the corpus — that is the self-dispatching queue chain (`chain=true`). Not for local iteration either: build nub locally and run `harness\u002Fsearch.mjs` directly against a `\u002Ftmp` fixture when the question does not need a specific OS.\n\n**Related:** `harness\u002Fdivergence-scan.mjs` finds packages whose grant disagrees across OSes and ranks them — the usual source of the hypothesis you are about to test.\n","# Re-measuring a package in the build-jail corpus\n\nThe loop is: **edit → push → dispatch a debug run → watch → read the CELL LOGS**. Each step has a trap that has already cost a run.\n\nCorpus repo: `nubjs\u002Fbuild-jail-corpus` (public). Local clone: `~\u002Fnub-corpus-staging`. nub: branch `sandbox\u002Fintegration` at `~\u002F.cache\u002Fnub\u002Fworktrees\u002Finteg`.\n\n## 1. Push the harness change — the queue is the trap\n\n⛔ **ALWAYS `git commit -- \u003Cexplicit paths>`.** A bare `git commit` sweeps in whatever is staged, and `queue.ndjson` is almost always staged because reading the corpus stages it. A commit carrying a stale queue **releases rows other runners are actively measuring**. Measured: one such commit carried 194 lines of stale queue.\n\n```sh\ncd ~\u002Fnub-corpus-staging\ngit commit -q -m \"…\" -- harness\u002Fsearch.mjs          # explicit paths, always\ngit show --stat --oneline HEAD | head -5             # one second; catches exactly this\n```\n\nThe push races ~12 concurrent runners, so it needs a retry loop — and the cleanup before each attempt is what makes it work:\n\n```sh\nfor a in 1 2 3 4 5; do\n  git restore --staged . 2>\u002Fdev\u002Fnull                 # unstage (checkout STAGES; restore is what unstages)\n  git checkout -- records queue.ndjson 2>\u002Fdev\u002Fnull   # drop working-copy edits — origin owns these\n  git clean -fdq -- records 2>\u002Fdev\u002Fnull              # reading the corpus leaves UNTRACKED records behind\n  git fetch -q origin main\n  git rebase -q origin\u002Fmain >\u002Fdev\u002Fnull 2>&1 || { git rebase --abort 2>\u002Fdev\u002Fnull; sleep 3; continue; }\n  git push origin HEAD:main >\u002Fdev\u002Fnull 2>&1 && { echo \"PUSHED\"; break; }\n  sleep 3\ndone\ngit fetch -q origin main\ngit merge-base --is-ancestor HEAD origin\u002Fmain && echo \"✓ LANDED\" || echo \"⛔ NOT LANDED\"\n```\n\n⛔ **Verify it landed. Never infer it from the push command's apparent success** — read `merge-base --is-ancestor`, because a rejected push inside a loop is easy to miss.\n\n**Reading the corpus for analysis leaves two kinds of debris**, both of which break the next rebase: `git checkout origin\u002Fmain -- records` STAGES those files, and it leaves records your HEAD lacks as UNTRACKED. Before cleaning untracked records, confirm they exist at origin (`git cat-file -e origin\u002Fmain:\u003Cpath>`) so a clean can never destroy an unpushed measurement.\n\n## 2. Dispatch the re-measure\n\nThe workflow has a **`packages` debug input**. When set, the runner SKIPS the queue entirely — no claim, no completion, no commit of results — and measures exactly those specs with `--force`, publishing only the ARTIFACT. That is what makes it safe to run against a live corpus.\n\n```sh\nkeyward --reason \"\u003Cthe task>\" -- gh workflow run corpus-queue-runner.yml -R nubjs\u002Fbuild-jail-corpus \\\n  -f os=macos \\                       # linux | macos | windows\n  -f nub_sha=\u003C40-char sha> \\\n  -f chain=false \\                    # ⛔ or it self-dispatches a whole chain\n  -f slice=2 \\\n  -f packages=\"husky@3.1.0 detox@20.9.1\"\n```\n\n- ⛔ **`nub_sha` must be PUSHED.** CI clones `nubjs\u002Fnub` anonymously. Check with `git ls-remote https:\u002F\u002Fgithub.com\u002Fnubjs\u002Fnub sandbox\u002Fintegration` before dispatching, or the run dies at clone.\n- **Pick the sha deliberately.** Using the sha the fleet already built gets a binary-cache hit (~0 min instead of a ~26 min cold Windows build) AND makes the result comparable with corpus records. Use a NEW sha only when testing a nub change.\n- **Platform speed differs enormously.** macOS\u002FLinux ≈ 1 min\u002Fpackage; Windows ≈ 9 min\u002Fpackage average and up to ~40 for a package that walks the full 55-cell ladder, plus ~26 min for a cold build. **Prototype on macOS or Linux whenever the question is not Windows-specific.**\n- ⛔ **In `run-batch.sh`, `--force` must come BEFORE `--file`** — it is parsed only as the first argument after `\u003Cnub>`. The workflow already does this. A run with the order wrong \"succeeded\" in 17 seconds having measured nothing.\n\n## 3. Watch it — and gate the watcher, or it will lie to you\n\n⛔ **NEVER dispatch CI without a bounded background watcher.** A stall is silent. And the watcher itself must gate, because **`gh run view --log` returns an EMPTY string for an in-progress run** — so `grep -q \u003Cbad-thing>` returns false and reports a PASS off nothing. That exact bug appeared inside a gate written to prevent it.\n\nThree gates, in order, all mandatory:\n\n1. **Terminal state**, on a BOUNDED loop that exits non-zero on timeout (`for i in $(seq 1 240); do … sleep 45; done`) so it can never hang forever.\n2. **Log non-empty** — `wc -l` over 100 lines, else refuse to interpret.\n3. **The canary RAN** — assert `fixture canary OK: \u003Cn> files` is PRESENT, never infer it from the absence of `REFUSING TO RUN`.\n\nThe canary is the harness's own guard: it installs `puppeteer@25.4.0` and refuses the whole batch unless the fixture produces >5,000 files. **If it refuses, nothing below it means anything** — that is how a broken isolation change gets caught before it writes garbage.\n\n## 4. Read the CELL LOGS, not just the grant\n\n⛔ **THE SINGLE MOST EXPENSIVE LESSON HERE.** Three consecutive fixes for one Windows defect were wrong because each was theorised from a grant and a path list. The failing cell's log named the exact cause in one line, every time.\n\nPer-cell logs are gitignored, so they exist ONLY in the run artifact — and only when the batch actually ran:\n\n```sh\ngh run download \u003Crun-id> -R nubjs\u002Fbuild-jail-corpus -D \u002Ftmp\u002Fart -n records-\u003Cos>-\u003Crun-id>\nfind \u002Ftmp\u002Fart -path '*\u003Cpkg>*' -name '*.log'     # one log per cell, named for the state\n```\n\nRead the cell **one rung below the winner** — that is where the reason lives. Read it RAW before grepping: a grep that matches nothing looks identical to a clean run. Real examples, each of which redirected an investigation:\n\n| log line | what it actually meant |\n|---|---|\n| `spawnSync git EPERM` | needs to EXECUTE git — an exec denial, not a write denial |\n| `mkdir '…\\Packages\\nub_sbx_\u003Cpid>_\u003Cnonce>_0'` | the leaf is the PROFILE dir, not `Packages` — a fixture cannot pre-create it |\n| `Could not find JAVA, skipping …` | a real line that was the WRONG explanation — both platforms skipped identically |\n| *(no error at all)* | a silent partial install; the cell failed on DIGEST, not exit code |\n\n⛔ **A record's `grant` tells you WHAT was needed; only the log tells you WHY.** And a plausible log line can still be the wrong explanation — check whether it differs across the platforms that disagree.\n\n## 5. Attribute the result\n\n- **`corpusGitSha`** in `provenance` names the corpus commit exactly. Use it — `harnessSha256` is a content hash and git rewrites line endings on Windows, so the same commit hashes differently there and **cannot be compared across platforms**.\n- **Confounds before blaming a platform:** same `nubGitSha`? same `nodeSelection.chosenMajor`? And note darwin records are arm64 while linux records are x64, so OS and ARCH move together and cannot be separated from these records alone.\n- **A debug run writes NOTHING to the corpus** — results live only in the artifact. To make a re-measurement stick, let the fleet re-claim the row (reopen it) rather than hand-editing records.\n\n## Scope\n\n**Use it for** a hypothesis about one package, verifying a harness or nub fix on a real runner, or getting cell logs for a package whose artifact expired (30-day retention).\n\n**Not for** running the corpus — that is the self-dispatching queue chain (`chain=true`). Not for local iteration either: build nub locally and run `harness\u002Fsearch.mjs` directly against a `\u002Ftmp` fixture when the question does not need a specific OS.\n\n**Related:** `harness\u002Fdivergence-scan.mjs` finds packages whose grant disagrees across OSes and ranks them — the usual source of the hypothesis you are about to test.\n",{},"main","9b734566fd36ceed8379f67a92291bc68af60801",".claude\u002Fskills\u002Fcorpus-remeasure",71,"https:\u002F\u002Fgithub.com\u002Fnubjs\u002Fnub","https:\u002F\u002Fnubjs.com","https:\u002F\u002Fgithub.com\u002Fnubjs","The all-in-one Node.js toolkit","2026-11-14 14:11:35",{"source":7,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add nubjs\u002Fnub --skill corpus-remeasure",{"tool":38,"command":39},"bunx","bunx skills add nubjs\u002Fnub --skill corpus-remeasure",{"tool":41,"command":42},"pnpm","pnpm dlx skills add nubjs\u002Fnub --skill corpus-remeasure","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fnubjs\u002Fnub\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fcorpus-remeasure\u002FSKILL.md",true,[],[47,53,59,66,73,80],{"skillNo":48,"name":49,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":50,"categories":51,"keywords":52,"publishedAt":18},"01M758BSH88RVXBG8TPCT427TY","nub","https:\u002F\u002Fgithub.com\u002Fnubjs\u002Fnub\u002Fblob\u002Fmain\u002Fskills\u002Fnub\u002FSKILL.md",[],[],{"skillNo":54,"name":49,"nameZh":6,"description":55,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":58,"publishedAt":18},"01M758BKPRHXS5MS27WG908PBG","This project uses nub, a single Rust CLI that runs on the project's installed Node and replaces the everyday Node toolchain. ALWAYS prefer nub over node, bun, npm, npx, pnpm, and yarn for: executing JS\u002FTS files, running package.json scripts, the npx-equivalent, installing dependencies, watch mode, and Node version management. Surfaces whenever you would otherwise reach for node\u002Fbun\u002Fnpm\u002Fnpx or a package manager.","https:\u002F\u002Fgithub.com\u002Fnubjs\u002Fnub\u002Fblob\u002Fmain\u002Fsite\u002Fpublic\u002Fskill.md",[],[],{"skillNo":60,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":18},"01M758BCR8A21H8P4TQ0YRHTC7","worktree","Create and manage isolated git worktrees for parallel build\u002Ftest\u002Flanding work on the nub repo. Invoke (via the Skill tool) whenever you need a fresh worktree to land a change, when you want to know what `.worktreeinclude` does or how to add an entry, or when cleaning up after a merge. Encodes the one-command setup (`nub scripts\u002Fnew-worktree.ts \u003Cslug>` or `node …`) that bakes in the proven recipe — worktree off origin\u002Fmain (vendor\u002Faube is plain in-tree files now, no submodule init), the shared CARGO_TARGET_DIR (`~\u002F.cache\u002Fnub\u002Fshared-target`) that all worktrees reuse, and applying `.worktreeinclude` — plus the eagerly-pull-the-shared-tree discipline and the safe cleanup path. Pairs with the `dev-loop` build skill.","https:\u002F\u002Fgithub.com\u002Fnubjs\u002Fnub\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fworktree\u002FSKILL.md",[],[],{"skillNo":67,"name":68,"nameZh":6,"description":69,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":70,"categories":71,"keywords":72,"publishedAt":18},"01M758B9NZ1GQMD1A6B25NK0D9","windows-vm-test","Run ad-hoc Nub tests and debugging probes in the maintainer's existing local Windows 11 ARM64 QEMU\u002FHVF VM when its external disk and runbook are present. Use for interactive win32-arm64 or Windows-on-ARM x64-emulation checks; use ci-adhoc-test for true x64-native behavior or when the local VM is unavailable.","https:\u002F\u002Fgithub.com\u002Fnubjs\u002Fnub\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fwindows-vm-test\u002FSKILL.md",[],[],{"skillNo":74,"name":75,"nameZh":6,"description":76,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":18},"01M758B2JHMBB6YJTGYSJR65S4","visual-review","Verify UI\u002Flayout\u002Fstyling changes are correct by computing occlusion, clipping, and alignment from the browser's resolved paint order via the chrome-devtools MCP `evaluate_script` tool — instead of eyeballing a flat screenshot. Invoke BEFORE declaring any UI, site, or styling\u002Flayout change correct. Screenshots have no depth buffer, so z-index\u002Focclusion\u002Fclip bugs are exactly where \"just look at it\" fails; the `evaluate_script` routines below turn those fuzzy visual judgments into deterministic measurements — including optical center-of-mass, which measures the glyph ink's true visual center so differently-sized labels can be aligned by more than eye.","https:\u002F\u002Fgithub.com\u002Fnubjs\u002Fnub\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fvisual-review\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":7,"ownerLogin":8,"ownerAvatarUrl":9,"ownerAccountNo":10,"ownerDisplayName":11,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":18},"01M758AWPM7CE440NBAYJHD44J","type-declarations","Add, audit, or update @nubjs\u002Ftypes declarations for Nub runtime APIs. Use whenever Nub adds or changes a user-visible global, built-in method, module loader type, or TC39 proposal; when TypeScript or @types\u002Fnode gains an overlapping declaration; and before every release to reconcile runtime changes since the previous tag. Covers conflict-safe global augmentation, compiler-version routing, fixture matrices, and package verification.","https:\u002F\u002Fgithub.com\u002Fnubjs\u002Fnub\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ftype-declarations\u002FSKILL.md",[],[]]