perf(release): trim Windows packaging setup

[t3dotgg]

Windows arm64 is the slowest release job (532s in the last nightly), and Windows x64 takes about 390s. Much of their setup is slow file work on Windows. Three changes to release-desktop.yml, based on recent nightlies and on timing runs on the same runners (windows-11-arm, blacksmith-32vcpu-windows-2025) via a scratch copy of the manual Windows test workflow.

Parallel checkout. git -c checkout.workers=0 checkout writes the files on every core. It applies to every desktop job (Linux and macOS too), but I only benchmarked Windows. On Linux the whole checkout already takes under a second of file writes. Nightly checkouts take 33-43s on arm64 and 54-80s on x64. To rule out order and machine effects, I ran both ways alternately, three times each, in one job (run):

Job	Without	With checkout.workers=0
arm64 a	34.6s	19.1s
arm64 b	23.0s	13.0s
x64 a	25.3s	23.6s
x64 b	33.0s	22.0s

A blobless fetch was slower on both runners, so the fetch stays as it is.

No Windows package cache. On Windows, linking the packages takes most of the install time, not downloading them. So restoring the ~700 MB pnpm store does not pay for itself. From the last ten nightlies:

	Cache hit: restore + install	Cache miss: registry install
arm64	183-233s (mean ~212s)	142-210s (mean ~180s), then ~30s to save the cache
x64	116-172s (mean ~134s)	84-160s (mean ~118s)

The arm64 entries also took 4.09 GB of this repository's Actions cache, which is over its 10 GB limit and evicting other entries. Linux and macOS keep the setup-vp cache.

Expect Windows arm64, the release's longest job, to finish about 70s sooner on average, and Windows x64 about 60s sooner. The benchmark varies a lot between runs, especially on x64, so a few nightlies will show the real average.

Skip the Spectre libraries when the resource monitor is cached. Only the Rust resource monitor build needs the Spectre-mitigated MSVC libraries. With a cached monitor, the desktop build's preflight (scripts/build-desktop-artifact.ts) already skips its MSVC check. node-pty uses its prebuilt Windows binaries, and the last nightly's Windows logs show no node-gyp, MSBuild, or cl.exe run. So the Visual Studio Installer step, 30-64s in recent nightlies, now runs only on a monitor cache miss.

I also tried running the Visual Studio Installer (MSVC Spectre libs) in the background during the package install. It did not survive the end of the step that started it, and running it at the same time slowed the install, so it is not in this PR.

Created with Claude Opus 5.5 in Claude Code, running inside T3 Code.

🤖 Generated with Claude Code

[macroscopeapp]

Approvability

Verdict: Approved at 1704f3b

Macroscope's review found this PR approvable — This is a narrowly scoped release-workflow optimization that changes Windows runner caching and setup timing while preserving the same packaging targets and artifact flow. It does not change product behavior, defaults, schemas, deployment targets, or static-analysis settings.

No code changes detected at fa5a7e8. Prior analysis still applies.

^{You can add or adjust custom eligibility rules. Learn more.}

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	13.5 KiB	13.5 KiB	+8 B (+0.1%)	15.1 KiB	✅
Codex	Thread snapshot wire	7.1 KiB	7.1 KiB	+1 B (+0.0%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.5 KiB	6.5 KiB	+7 B (+0.1%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	56.3 KiB	56.3 KiB	0 B (0.0%)	66.4 KiB	✅
Codex	Live turn messages	10	10	0 (0.0%)	21	✅
Claude	Total thread wire	13.5 KiB	13.5 KiB	−3 B (−0.0%)	15.1 KiB	✅
Claude	Thread snapshot wire	7.1 KiB	7.1 KiB	+5 B (+0.1%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.4 KiB	6.4 KiB	−8 B (−0.1%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	57.0 KiB	57.0 KiB	0 B (0.0%)	66.4 KiB	✅
Claude	Live turn messages	9	9	0 (0.0%)	21	✅

Baseline: 6f8e253 · PR result: fa5a7e8 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 114.0 KiB
• Claude decoded thread snapshot: 114.7 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.

[coderabbitai]

Navigate logical layers of code changes, visualize relationships, and explore their blast radius.

Important

Review skipped

Review was skipped as selected files did not have any reviewable changes.

⚙️ Run configuration

Configuration used: Repository: pingdotgg/t3code/.coderabbit.yaml

Review profile: CHILL

Plan: Team

Run ID: dcc2369c-b96e-45d7-8cda-55568327802d

📥 Commits

Reviewing files that changed from the base of the PR and between 1704f3b and fa5a7e8.

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Repository: pingdotgg/t3code/.coderabbit.yaml

Review profile: CHILL

Plan: Team

Run ID: eda0c89e-aa8c-45bd-b7a5-ed503077927a

📥 Commits

Reviewing files that changed from the base of the PR and between b2982fa and 1704f3b.

📒 Files selected for processing (1)

• .github/workflows/release-desktop.yml

Included review availability: This review used your included allowance. Your plan provides up to 10 included reviews per hour; 4 remain after this review.

---

📝 Walkthrough

Walkthrough

The desktop release workflow sets checkout.workers=0 for its detached checkout and removes Windows package-cache path resolution and cache steps. It also limits Spectre-mitigated MSVC library installation to Windows jobs where the resource-monitor cache misses.

Changes

Desktop release workflow

Layer / File(s)	Summary
Windows release workflow setup .github/workflows/release-desktop.yml	The workflow documents checkout observations and sets checkout.workers=0. It removes Windows package-cache path resolution and cache steps. It installs Spectre-mitigated MSVC libraries only on Windows jobs where the resource-monitor cache misses.

Priority: ⬇️ Low

Estimated code review effort: 2 (Simple) | ~8 minutes

Change: Refactor

Suggested reviewers: juliusmarminge

Merge Risk: ⚪ Minimal · up to 1704f

The workflow changes have no established user-facing failure, and the conditional Spectre setup does not block the cache-hit build path. The PR is mergeable subject to normal checks.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly and concisely describes the main change: reducing Windows release packaging setup time.
Description check	✅ Passed	The description provides detailed problem context, implementation changes, benchmark results, limitations, and agent attribution. It does not include an explicit Scope and approval section or link to …
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 0…
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Commit to this branch
• Create a new PR

---

_{Comment @coderabbitai help to get the list of available commands.}

Dismissing prior approval to re-evaluate b2982fa

Dismissing prior approval to re-evaluate 1704f3b