[prisis]

Summary

• vis test runs the package's test script when one exists, through the detected package manager. A warning is printed first. Otherwise it runs Vitest, or Rstest when Vitest is not installed. Both are found by walking up node_modules.
• No runner installed: vis test asks whether to add Vitest or Rstest as a devDependency, then runs it. Non-interactive runs stop with an error that lists the install commands. Cancelling installs nothing.
• --watch is rejected when a test script is used, so the flag is never dropped silently.
• vis check is now a gate: it runs vis lint, then vis fmt --check, over the same scope. The type check runs when TypeScript is detected, unless lint.adapters.typecheck.enabled is false.
• vis outdated is the catalog dependency check, previously vis check.

Breaking change

vis check no longer reports outdated dependencies. Scripts and CI that use vis check for that now need vis outdated. Old flags on vis check (--exit-code, --security, --format json) fail as unknown options. vis check <package> fails with a message that points to vis outdated. The c alias is removed.

Testing

• Typecheck and ESLint are clean.
• The native addons (vis, secret-scanner, task-runner) build locally, and the command, util, and inference suites pass: 214 files, 2794 tests.
• vis check was run against the built CLI in sandboxes: exit 1 on a type error, exit 0 on a clean project, and a clear error for vis check react.
• vis test was run against the built CLI: it lists the runners and installs nothing when non-interactive. The Rstest fallback forwards arguments and propagates the exit code.

[github-actions]

Thank you for following the naming conventions! 🙏

[pkg-pr-new]

Open in StackBlitz

npm i https://pkg.pr.new/@visulima/vis@34

npm i https://pkg.pr.new/@visulima/vis-mcp@34

commit: 1b11bf9

[codspeed]

Merging this PR will not alter performance

⚠️ 8 benchmarks spent significant time in system calls

System calls cannot be consistently instrumented, so they are not included in the measure, which understates the real cost. Please switch to the Walltime instrument to accurately measure system calls.

Measurement and system calls

• detectLockfileDrift · aube over foreign lockfile
• resolveInstaller · explicit backend (no fs probe)
• resolveLocalBin · miss (full ancestor walk → PM fallback)
• detectPm · npm workspace
• resolveInstaller · auto (detect → pnpm)
• × 2800 packages
• detectPm · pnpm workspace
• detectPm · bun workspace

✅ 12 untouched benchmarks
⏩ 101 skipped benchmarks¹

---

_{Comparing t3/node-runtime-feature-parity (1b11bf9) with main (c71b539)}

Footnotes

1. 101 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩