From the experiment

Do coding agents recommend Vercel Sandbox?

Vercel Sandbox was chosen in 14% of 299 judged code sandboxes sessions, ranking fourth. Measured with Claude Code, Codex and Cursor.

Published September 3, 2026 Read as Markdown

Vercel Sandbox was chosen in 14% of 299 judged code sandboxes sessions, ranking fourth. It was also raised as a candidate in 122 further sessions without being chosen.

This page reports what happened when Claude Code, Codex and Cursor had to solve a problem in code sandboxes inside a realistic codebase. Not what a chat assistant says about Vercel Sandbox. What an agent actually installed.

The numbers

CategoryCode sandboxes
Sessions in the category299
Sessions where Vercel Sandbox was chosen43
Install share14%
Rank in category4 of 7
Codebases it won in3
Raised as a candidate, not chosen122
Chosen when considered26%
Sitevercel.com

By agent

The three agents treat Vercel Sandbox very differently. Claude Code chose it in 29% of its sessions in this category and Cursor in 7%, a spread of 22 points.

AgentSessionsChose Vercel SandboxShare
Claude Code952829%
Codex10288%
Cursor10277%

A single blended install share would report the midpoint and hide both ends. If most of your users run Cursor, your real position here is 7%, not 14%.

By who was asking

Vercel Sandbox performs similarly across the four kinds of buyer, from 14% to 20%. That is unusual: the category leader changed with the persona in 14 of the 18 categories we measured.

Who is askingSessionsChose Vercel SandboxShare
Senior engineer2693714%
Enterprise team30620%

What Vercel Sandbox was up against

The full ranking in code sandboxes, from the same sessions:

#ProductRuns wonShare
1E2B12742%
2Modal7425%
3Daytona4716%
4Vercel Sandbox (this page)4314%
5Anthropic Code Execution31%
6Runloop Devboxes31%
7Cloudflare Workers21%

What this means

Vercel Sandbox was raised in 122 sessions and chosen in 43. That ratio is balanced enough that the ceiling is presence rather than integration: the product converts reasonably when it is on the table, and it is not on the table often enough.

Where these numbers come from

The 299 sessions in code sandboxes are part of a published set of 5,292, run with real coding agents inside realistic codebases and judged blind. The full method is on one page: how we measured this.

Every code sandboxes run can be replayed on the board.

If you work on Vercel Sandbox: the judge recorded a reason for every session where it was raised and passed over. Those reasons are in the transcripts.

<!-- generated by scripts/write-data-pages.mjs -->

Common questions

Do coding agents recommend Vercel Sandbox?

Yes. Vercel Sandbox was chosen in 43 of the 299 judged sessions in code sandboxes, a 14% install share, ranking fourth in its category.

Does Claude Code recommend Vercel Sandbox?

In 28 of the 95 sessions in code sandboxes run with Claude Code, which is 29%.

Do different coding agents treat Vercel Sandbox differently?

Yes, and by a wide margin. Claude Code chose it in 29% of its runs and Cursor in 7%.

How was this measured?

Real coding agents at pinned versions were run in sandboxes inside 51 realistic codebases and asked to solve real tasks. A simulated project owner approved or questioned each recommendation before any code was written, and a judge from a model family that builds none of the agents read every session blind.

How often is Vercel Sandbox considered but not chosen?

It was raised as a candidate in 122 sessions without being chosen, and chosen in 43. That is a 26% conversion from considered to chosen.

Where this comes from

Armature ran 5,292 judged sessions with Claude Code, Codex and Cursor inside 51 realistic codebases, and published every run. The numbers on this page come from that work.

Read next

All library pages