feat: test failure scores 0 + set vercel pkg to 1.7.15

[kisernl]

No description provided.

[github-actions]

Storage Benchmark Results

10MB Files

#	Provider	Score	Download	Throughput	Upload	Status
1	Tigris	93.7	0.12s	704.9 Mbps	1.02s	10/10
2	AWS S3	93.2	0.12s	686.1 Mbps	0.39s	10/10
3	Cloudflare R2	82.0	0.49s	171.6 Mbps	0.98s	10/10

---

View full run · SVGs available as build artifacts

[github-actions]

Sandbox Benchmark Results

Sequential

#	Provider	Score	Median TTI	P95	P99	Status
1	daytona	98.0	0.14s	0.29s	0.29s	10/10
2	e2b	94.8	0.44s	0.64s	0.64s	10/10
3	blaxel	93.3	0.49s	0.94s	0.94s	10/10
4	vercel	82.0	1.76s	1.87s	1.87s	10/10
5	runloop	80.8	1.87s	2.00s	2.00s	10/10
6	codesandbox	74.3	2.36s	2.89s	2.89s	10/10
7	cloudflare	73.7	2.27s	3.18s	3.18s	10/10
8	modal	68.5	2.10s	2.82s	2.82s	9/10
9	namespace	65.4	2.07s	5.54s	5.54s	10/10
10	hopx	53.2	1.13s	15.75s	15.75s	10/10
11	just-bash	0.0	0.00s	0.00s	0.00s	0/10

Staggered

#	Provider	Score	Median TTI	P95	P99	Status
1	daytona	98.6	0.12s	0.18s	0.18s	10/10
2	blaxel	94.3	0.49s	0.69s	0.69s	10/10
3	e2b	93.5	0.55s	0.81s	0.81s	10/10
4	hopx	88.3	1.01s	1.40s	1.40s	10/10
5	modal	78.2	1.92s	2.58s	2.58s	10/10
6	runloop	76.7	1.96s	2.89s	2.89s	10/10
7	vercel	76.5	1.74s	3.25s	3.25s	10/10
8	namespace	75.5	2.24s	2.75s	2.75s	10/10
9	codesandbox	72.5	2.62s	2.95s	2.95s	10/10
10	cloudflare	72.0	2.38s	3.44s	3.44s	10/10
11	just-bash	0.0	0.00s	0.00s	0.00s	0/10

Burst

#	Provider	Score	Median TTI	P95	P99	Status
1	daytona	97.5	0.18s	0.37s	0.37s	10/10
2	blaxel	93.2	0.63s	0.77s	0.77s	10/10
3	e2b	93.2	0.51s	0.95s	0.95s	10/10
4	hopx	83.8	1.59s	1.66s	1.66s	10/10
5	vercel	81.4	1.71s	2.09s	2.09s	10/10
6	runloop	81.0	1.87s	1.95s	1.95s	10/10
7	cloudflare	76.1	2.34s	2.47s	2.47s	10/10
8	namespace	76.0	2.33s	2.51s	2.51s	10/10
9	modal	74.7	2.26s	2.95s	2.95s	10/10
10	codesandbox	69.0	2.86s	3.46s	3.46s	10/10
11	just-bash	0.0	0.00s	0.00s	0.00s	0/10

---

View full run · SVGs available as build artifacts

[Copilot]

Pull request overview

Updates benchmark result handling so provider runs with 0 successful iterations are treated as real failures (score = 0) rather than being marked as “skipped”, and pins the Vercel provider dependency to a specific version.

Changes:

• Remove “all iterations failed ⇒ skipped” behavior across sandbox/storage/browser benchmarks and result renderers.
• Ensure stats/time-to-first-ready calculations don’t error on empty successful sets (default to zeros).
• Pin @computesdk/vercel to 1.7.15 (and update lockfile accordingly).

Reviewed changes

Copilot reviewed 9 out of 10 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
src/storage/benchmark.ts	Stops marking all-failure storage runs as skipped; computes stats from successful-only set.
src/browser/benchmark.ts	Stops marking all-failure browser runs as skipped; stats already handle empty successful set.
src/sandbox/benchmark.ts	Removes skip-on-all-failures and defaults stats to zeros when no successes.
src/sandbox/concurrent.ts	Removes skip-on-all-failures; guards min/stats when no successes.
src/sandbox/staggered.ts	Removes skip-on-all-failures; guards min/stats when no successes.
src/sandbox/table.ts	Removes special FAILED row; always reports success ratio in “Status”.
src/run.ts	Adds (ok/total OK) to printed scores for storage and browser modes.
src/merge-results.ts	Removes special FAILED rows in merged storage/browser tables; always reports success ratio.
package.json	Pins @computesdk/vercel to 1.7.15.
package-lock.json	Updates lockfile to match the pinned @computesdk/vercel version.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.