Skip remote npm cache on VPS runners

[nish3451]

Summary

• keep setup-node npm caching on GitHub-hosted fallback runners
• disable the redundant 1.06 GiB Actions cache on persistent self-hosted VPS runners
• preserve CI's no-lockfile fallback and add a workflow contract test

Why

The exact-main production preflight stalled for more than 15 minutes restoring the npm Actions cache, then hit an Azure cache authentication failure. The VPS runners already share a persistent local npm cache, so the remote archive is redundant and blocks evidence/deploy throughput.

Verification

• workflow YAML parses
• workflow cache contract passes via static assertion
• git diff --check clean
• sgscan clean
• CodeRabbit local gate: 0 findings
• independent code review: approved after preserving the CI no-lockfile guard

No deploy or provider mutation is included in this PR.

---

Note

Low Risk
Workflow-only changes to dependency caching; no application, auth, or deploy logic is modified.

Overview
Disables GitHub Actions npm caching on self-hosted runners while keeping it on github-hosted fallbacks, so VPS jobs no longer spend time restoring large remote cache archives that duplicate the machine’s persistent local npm store.

Across CI, deploy, D1 backup/restore, cross-browser matrix, and production soak workflows, actions/setup-node@v6 now sets cache and package-manager-cache only when runner.environment == 'github-hosted'. CI’s codex-node-checks job still ties caching to a present package-lock.json so the no-lockfile npm install path stays unchanged.

Adds tests/self-hosted-node-cache-workflows.test.ts to assert runner-routed jobs use the hosted-only cache expressions (including the CI lockfile guard).

^{Reviewed by Cursor Bugbot for commit 03fc2f9. Bugbot is set up for automated code reviews on this repo. Configure here.}

Summary by CodeRabbit

• Bug Fixes
  • Updated Node.js dependency caching in CI and deployment workflows to run only on GitHub-hosted runners.
  • Disabled cache configuration on non-hosted runners to avoid unsupported caching behavior.
  • Added/standardized package-manager-cache settings alongside runner-gated cache values, with lockfile-aware caching where applicable.
• Tests
  • Added automated coverage to verify runner-specific actions/setup-node caching configuration across selected workflows.

[greptile-apps]

nish3451 has reached the 50-credit limit for trial accounts. To continue receiving code reviews, upgrade your plan.

[cursor]

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_57ac10c0-c498-42f6-9660-60bbec790cbb)

[coderabbitai]

📝 Walkthrough

Walkthrough

The workflows now enable npm caching only on GitHub-hosted runners, with lockfile-aware conditions in CI. A Vitest test parses routed workflows and verifies runner selection, setup-node cache inputs, and the expected count of configured setup-node steps.

Changes

Hosted runner cache configuration

Layer / File(s)	Summary
Conditional setup-node caching .github/workflows/*.yml	Setup-node cache and package-manager-cache inputs now use GitHub-hosted-only expressions; CI additionally requires a non-empty lockfile hash.
Workflow cache assertions tests/self-hosted-node-cache-workflows.test.ts	Adds workflow routing, typed setup-node data, and assertions for runner expressions, cache expressions, and the nine configured setup-node steps.

Estimated code review effort: 2 (Simple) | ~10 minutes

Possibly related PRs

• nish3451/0509#425: Modifies the same D1 workflow setup-node cache inputs with conditional npm caching.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
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.
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly summarizes the main change: disabling remote npm cache usage on VPS/self-hosted runners.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch fix/self-hosted-cache

---

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

[coderabbitai]

Actionable comments posted: 1

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@tests/self-hosted-node-cache-workflows.test.ts`:
- Line 55: Update the runnerRoutedJobs count assertion in the self-hosted node
cache workflow test to expect the 11 jobs produced by runnerRoutedWorkflows,
unless the missing twelfth job is intentionally required and should be added to
that workflow definition.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 90e6da13-f805-44ba-9d13-915be07f30a7

📥 Commits

Reviewing files that changed from the base of the PR and between 2558ffe and c37e8ff.

📒 Files selected for processing (8)

• .github/workflows/ci.yml
• .github/workflows/cross-browser-matrix.yml
• .github/workflows/d1-backup-r2.yml
• .github/workflows/d1-backup-validate.yml
• .github/workflows/d1-remote-restore-evidence.yml
• .github/workflows/deploy-production.yml
• .github/workflows/finalize-production-soak.yml
• tests/self-hosted-node-cache-workflows.test.ts

[greptile-apps]

nish3451 has reached the 50-credit limit for trial accounts. To continue receiving code reviews, upgrade your plan.

[cursor]

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_50037e88-ce73-428e-ad23-92031c9ab954)