feat: performance benchmarking suite, CI comment report, and template tsconfig fix

[Copilot]

Implements the performance benchmarking suite from the implementation plan, wires it into CI with a PR comment report mirroring the coverage-report pattern, and fixes a long-standing root tsconfig.json gap that caused ~200 false-positive type errors.

Benchmarks

Five Vitest bench suites under benchmarks/:

File	What's measured
renderer/ring-buffer.bench.ts	SharedRingBuffer write/read at 64 B / 1 KB / 16 KB + sequential throughput + introspection
renderer/shared-buffer.bench.ts	SharedBuffer allocation + JSON payload round-trips at ~100 B / ~2 KB / ~10 KB
utils/safe-redirect.bench.ts	safeRedirect valid paths, rejection fast-paths, percent-encoded bypass attempts
utils/promise.bench.ts	promiseHash (2–10 entries) + timeout wrapper overhead
utils/ip-address.bench.ts	getClientIPAddress across all header patterns including RFC 7239 Forwarded and multi-hop x-forwarded-for

New scripts in package.json:

pnpm bench       # interactive / watch mode
pnpm bench:ci    # single run, writes benchmarks/results.json

A dedicated vitest.bench.config.ts keeps bench files out of pnpm test.

CI workflow (.github/workflows/benchmarks.yml)

Two jobs, same structure as coverage-report:

• benchmark — runs pnpm bench:ci on Node 24, uploads benchmark-results artifact (30-day retention). Fires on push to main and on PRs touching benchmarks/**, packages/**, vitest.bench.config.ts, or pnpm-lock.yaml.
• benchmark-report — PR-only. Downloads current + base (main) artifacts via actions/download-artifact + dawidd6/action-download-artifact, then uses actions/github-script to post (or update) a single comment. Uses a hidden HTML marker to avoid comment spam on re-runs.

Comment format — one collapsible <details> block per file:

Suite — Benchmark	ops/sec	mean (µs)	p99 (µs)	±rme	vs main
SharedRingBuffer – small payload (64 B) — write 64 bytes	545.5 K	0.0018	0.0072	±0.54%	-13.0% ⚠️
SharedRingBuffer – medium payload (1 KB) — write 1 KB	295.1 K	0.0034	0.0128	±0.53%	+13.6% 🚀

Delta thresholds: ⚠️ = >10% slower, 🚀 = >10% faster. Delta column is omitted on the first run (no base artifact on main yet).

Docs

• docs/benchmarks.md — reference page covering all suites, output columns, config, and how to add a new benchmark
• VitePress sidebar updated under "Testing & Tooling"
• README.md gets a Benchmarks section
• IMPLEMENTATION_PLAN.md item marked done

tsconfig fix

packages/create-scratchy-app/src/template/ imports external packages (@builder.io/qwik, drizzle-orm, etc.) that aren't installed in this monorepo — they're dependencies of the generated app. The per-package tsconfig already excluded src/template; the root one didn't, producing ~200 spurious errors when running tsc --noEmit directly.

// tsconfig.json — added to exclude array
"packages/create-scratchy-app/src/template"

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

[socket-security]

Review the following changes in direct dependencies. Learn more about Socket for GitHub.

Diff	Package	Supply Chain Security	Vulnerability	Quality	Maintenance	License
	wrangler@​4.76.0

View full report

[github-actions]

📚 Docs preview deployed

https://scratchyjs-docs-pr-36.asjas.workers.dev

Updates automatically on every push to this PR.

[github-actions]

Coverage Report

Status	Category	Percentage	Covered / Total
🔵	Lines	98.67% 🟰 ±0%	2007 / 2034
🔵	Statements	98.07% 🟰 ±0%	2094 / 2135
🔵	Functions	99.25% 🟰 ±0%	399 / 402
🔵	Branches	92.61% 🟰 ±0%	1104 / 1192

File Coverage

No changed files found.

Generated in workflow #299 for commit d36a04f by the Vitest Coverage Report Action

[github-actions]

⚡ Benchmark Results

benchmarks/renderer/ring-buffer.bench.ts

Suite — Benchmark	ops/sec	mean (µs)	p99 (µs)	±rme
SharedRingBuffer – small payload (64 B) — write 64 bytes	518.9 K	0.0019	0.0075	±1.44%
SharedRingBuffer – small payload (64 B) — write + read 64 bytes	467.0 K	0.0021	0.0078	±1.42%
SharedRingBuffer – medium payload (1 KB) — write 1 KB	244.4 K	0.0041	0.0142	±0.58%
SharedRingBuffer – medium payload (1 KB) — write + read 1 KB	210.0 K	0.0048	0.0215	±1.33%
SharedRingBuffer – large payload (16 KB) — write 16 KB	57.5 K	0.0174	0.0424	±1.56%
SharedRingBuffer – large payload (16 KB) — write + read 16 KB	83.8 K	0.0119	0.0420	±18.76%
SharedRingBuffer – sequential throughput (100 × 64 B) — 100 write + read cycles	40.7 K	0.0246	0.0441	±0.39%
SharedRingBuffer – introspection — availableToRead	14.47 M	0.0001	0.0001	±0.10%
SharedRingBuffer – introspection — isEmpty	14.88 M	0.0001	0.0001	±0.10%
SharedRingBuffer – introspection — isFull	14.74 M	0.0001	0.0001	±0.21%

benchmarks/renderer/shared-buffer.bench.ts

Suite — Benchmark	ops/sec	mean (µs)	p99 (µs)	±rme
SharedBuffer – allocation — createSharedBuffer(4 KB)	595.6 K	0.0017	0.0070	±1.15%
SharedBuffer – allocation — createSharedBuffer(64 KB)	43.3 K	0.0231	0.0426	±0.75%
SharedBuffer – small payload round-trip — write small JSON	314.4 K	0.0032	0.0120	±0.43%
SharedBuffer – small payload round-trip — write + read small JSON	246.4 K	0.0041	0.0135	±0.49%
SharedBuffer – medium payload round-trip — write medium JSON (~2 KB)	99.7 K	0.0100	0.0261	±0.40%
SharedBuffer – medium payload round-trip — write + read medium JSON (~2 KB)	57.1 K	0.0175	0.0390	±0.94%
SharedBuffer – large payload round-trip — write large JSON (~10 KB)	15.5 K	0.0647	0.0886	±0.31%
SharedBuffer – large payload round-trip — write + read large JSON (~10 KB)	7.7 K	0.1293	0.1658	±0.35%

benchmarks/utils/ip-address.bench.ts

Suite — Benchmark	ops/sec	mean (µs)	p99 (µs)	±rme
getClientIPAddress – no IP headers — no IP-related headers → null	2.55 M	0.0004	0.0006	±0.18%
getClientIPAddress – single header — cf-connecting-ip (Cloudflare)	1.68 M	0.0006	0.0010	±0.34%
getClientIPAddress – single header — x-forwarded-for (simple)	1.97 M	0.0005	0.0007	±0.28%
getClientIPAddress – single header — x-real-ip	1.47 M	0.0007	0.0011	±0.29%
getClientIPAddress – single header — true-client-ip (Akamai / Cloudflare Enterprise)	1.54 M	0.0006	0.0010	±0.19%
getClientIPAddress – x-forwarded-for multi-hop — 2-hop chain	1.78 M	0.0006	0.0008	±3.06%
getClientIPAddress – x-forwarded-for multi-hop — 4-hop chain	1.70 M	0.0006	0.0009	±1.13%
getClientIPAddress – Forwarded header (RFC 7239) — simple for= directive	991.0 K	0.0010	0.0015	±1.33%
getClientIPAddress – Forwarded header (RFC 7239) — for= with port	749.7 K	0.0013	0.0020	±0.35%
getClientIPAddress – Forwarded header (RFC 7239) — IPv6 literal	761.4 K	0.0013	0.0017	±0.27%
getClientIPAddress – Forwarded header (RFC 7239) — multi-hop Forwarded	834.8 K	0.0012	0.0017	±1.65%
getClientIPAddress – IPv6 addresses — x-forwarded-for IPv6	1.14 M	0.0009	0.0010	±0.28%

benchmarks/utils/promise.bench.ts

Suite — Benchmark	ops/sec	mean (µs)	p99 (µs)	±rme
promiseHash – concurrent resolution — 2 already-resolved promises	869.3 K	0.0012	0.0031	±1.03%
promiseHash – concurrent resolution — 5 already-resolved promises	524.5 K	0.0019	0.0034	±1.72%
promiseHash – concurrent resolution — 10 already-resolved promises	311.2 K	0.0032	0.0055	±1.57%
promiseHash – concurrent resolution — 5 promises with object values	485.4 K	0.0021	0.0026	±0.27%
timeout – wrapping fast promises — timeout wrapping an already-resolved promise (1 s budget)	965.2 K	0.0010	0.0015	±0.34%
timeout – wrapping fast promises — timeout wrapping an already-resolved object (5 s budget)	897.6 K	0.0011	0.0017	±0.31%

benchmarks/utils/safe-redirect.bench.ts

Suite — Benchmark	ops/sec	mean (µs)	p99 (µs)	±rme
safeRedirect – valid paths — root path /	6.89 M	0.0001	0.0002	±0.12%
safeRedirect – valid paths — simple path /dashboard	3.47 M	0.0003	0.0003	±0.11%
safeRedirect – valid paths — nested path /settings/profile	3.11 M	0.0003	0.0004	±0.10%
safeRedirect – valid paths — path with query string /search?q=hello	3.38 M	0.0003	0.0005	±0.10%
safeRedirect – valid paths — path with hash /docs#section	3.46 M	0.0003	0.0003	±0.10%
safeRedirect – rejected inputs — absolute URL https://evil.com	4.67 M	0.0002	0.0004	±0.10%
safeRedirect – rejected inputs — protocol-relative URL //evil.com	4.62 M	0.0002	0.0002	±0.09%
safeRedirect – rejected inputs — backslash-relative /\evil.com	4.63 M	0.0002	0.0002	±0.09%
safeRedirect – rejected inputs — path traversal /../etc/passwd	4.07 M	0.0002	0.0003	±0.11%
safeRedirect – rejected inputs — null input	16.04 M	0.0001	0.0001	±0.09%
safeRedirect – rejected inputs — undefined input	15.95 M	0.0001	0.0001	±0.11%
safeRedirect – rejected inputs — empty string	15.82 M	0.0001	0.0001	±0.17%
safeRedirect – percent-encoded bypass — percent-encoded // (%2F%2F)	4.40 M	0.0002	0.0004	±0.24%
safeRedirect – percent-encoded bypass — percent-encoded path traversal (%2e%2e)	4.79 M	0.0002	0.0002	±0.10%
safeRedirect – percent-encoded bypass — mixed percent-encoded absolute URL	4.45 M	0.0002	0.0003	±0.13%
safeRedirect – custom default redirect — valid path with custom default	3.56 M	0.0003	0.0003	±0.25%
safeRedirect – custom default redirect — invalid input with custom default	4.79 M	0.0002	0.0002	±0.10%

ℹ️ No base results from main found — delta column omitted on first run.

[Copilot]

Pull request overview

Adds a Vitest-powered benchmarking suite for key framework hot paths, wires it into CI with an automated PR comment report, and fixes root tsconfig.json exclusions to avoid template-related typecheck noise.

Changes:

• Added benchmarks/ suites plus a dedicated vitest.bench.config.ts and pnpm bench / pnpm bench:ci scripts.
• Introduced a new GitHub Actions workflow to run benchmarks and post/update a PR comment with results (and deltas vs main when available).
• Updated docs/README/sidebar and excluded the create-scratchy-app template directory from the root tsconfig.

Reviewed changes

Copilot reviewed 12 out of 14 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
vitest.bench.config.ts	Dedicated Vitest config to run .bench.ts files and emit JSON results.
package.json	Adds bench and bench:ci scripts for local and CI runs.
benchmarks/renderer/ring-buffer.bench.ts	Bench suite for SharedRingBuffer write/read throughput and getters.
benchmarks/renderer/shared-buffer.bench.ts	Bench suite for SharedBuffer allocation and JSON round-trips.
benchmarks/utils/safe-redirect.bench.ts	Bench suite for safeRedirect happy paths and rejection fast-paths.
benchmarks/utils/promise.bench.ts	Bench suite for promiseHash and timeout overhead.
benchmarks/utils/ip-address.bench.ts	Bench suite for getClientIPAddress across common header patterns.
.github/workflows/benchmarks.yml	New CI workflow to run benchmarks, upload artifacts, and post PR comment report.
docs/benchmarks.md	Reference documentation for running/adding benchmarks and interpreting output.
docs/.vitepress/config.ts	Adds Benchmarks page to the VitePress sidebar.
README.md	Adds a Benchmarks section and links to docs.
.gitignore	Ignores generated benchmarks/results.json.
tsconfig.json	Excludes create-scratchy-app template directory to avoid false-positive type errors.
IMPLEMENTATION_PLAN.md	Marks the benchmarking suite item complete.