benchmark-sandbox

benchmark-sandbox is a skill for Claude Code from build-with-dhiraj/ai-workflow-framework-portability-kit. It costs 51 tokens per session (5,749 once invoked), scanned B, original, MIT.

A benchmark runner that tests coding-agent skills inside temporary Vercel Sandboxes, which are short-lived virtual machines hosted by Vercel.

In plain words
What is it for?
Running multi-stage benchmark scenarios, checking skill injection across phases, scoring results as structured data, and producing coverage reports.
Why use it?
It keeps evaluation runs isolated from your computer and checks whether generated apps build, work through user stories, deploy, and produce expected results.

Skill for Claude Code

Written for Claude Code: PreToolUse hook event. Also seen: reads .claude/ paths; mentions Claude Code.

Not installable: its command points at a path on the author’s own machine, so it runs nowhere else. The line is /home/vercel-sandbox/.global/npm/bin/claude.

Part of the vercel-plugin plugin — 7 skills, 2 agents, 1 MCP server shipped together

Good fit Running multi-stage benchmark scenarios, checking skill injection across phases, scoring results as structured data, and producing coverage reports.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add build-with-dhiraj/ai-workflow-framework-portability-kit
Claude Code
/plugin install vercel-plugin

Made for: Claude Code.

Or install vercel-plugin, the plugin that ships this one along with the rest of its 7 skills, 2 agents, 1 MCP server.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for benchmark-sandbox

README.md
[![agentmods](https://agentmods.dev/badge/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/benchmark-sandbox/github.svg)](https://agentmods.dev/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/benchmark-sandbox)
Your own site
<a href="https://agentmods.dev/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/benchmark-sandbox"><img src="https://agentmods.dev/badge/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/benchmark-sandbox/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for benchmark-sandbox

Your own site · 80×15
<a href="https://agentmods.dev/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/benchmark-sandbox"><img src="https://agentmods.dev/badge/skills/build-with-dhiraj/ai-workflow-framework-portability-kit/benchmark-sandbox.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 51 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,749 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 2 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00051 $0.05749
Opus 5 $0.00026 $0.02874
Sonnet 5 $0.00010 $0.01150
Haiku 4.5 $0.00005 $0.00575

Measured 10d ago against content hash 2588bbda779e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade B, and why

benchmark-sandbox scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

The scan reads SKILL.md. This mod also ships 8 executable files (logger.ts, run-eval.ts, sandbox-analyze.ts, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

│ │ ├─ find ~/.claude/debug -type f (debug log count)

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

│ │ └─ curl localhost:3000 (port status)
Plugins/vercel-marketplace-source/.claude/skills/benchmark-sandbox/SKILL.md · 388 lines

How it starts

The opening of the file, as written. The whole thing — 388 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Benchmark Sandbox — Remote Eval via Vercel Sandboxes

Run benchmark scenarios inside Vercel Sandboxes — ephemeral Firecracker microVMs with node24. Each sandbox gets a fresh Claude Code + Vercel CLI + agent-browser install, the local vercel-plugin uploaded, and runs a 3-phase eval pipeline:

  • Phase 1 (BUILD): Claude Code builds the app with --dangerously-skip-permissions --debug
  • Phase 2 (VERIFY): A follow-up Claude Code session uses agent-browser to walk through user stories, fixing issues until all pass (20 min timeout)
  • Phase 3 (DEPLOY): A third Claude Code session links to vercel-labs, runs vercel deploy, and fixes build errors (up to 3 retries). Deployed apps have deployment protection enabled by default.

Skills are tracked across all 3 phases — each phase may trigger additional skill injections as new files/patterns are created. After each phase, a haiku structured scoring step (claude -p --json-schema --model haiku) evaluates the results as structured JSON.

Proven Working Script

Use run-eval.ts — the proven eval runner:

# Run default scenarios with full 3-phase pipeline
bun run .claude/skills/benchmark-sandbox/run-eval.ts

# With dynamic scenarios from a JSON file (recommended — see "Dynamic Scenarios" below)
bun run .claude/skills/benchmark-sandbox/run-eval.ts --scenarios-file /tmp/my-scenarios.json

# Keep sandboxes alive overnight with public URLs
bun run .claude/skills/benchmark-sandbox/run-eval.ts --keep-alive --keep-hours 8

# Build-only (skip verification and deploy)
bun run .claude/skills/benchmark-sandbox/run-eval.ts --skip-verify --skip-deploy

# Run specific scenarios by slug
bun run .claude/skills/benchmark-sandbox/run-eval.ts --scenarios splitwise-clone,calendly-clone

CLI Flags

Flag Default Description
--concurrency N 5 Max parallel sandboxes (max 10)
--timeout MS 1800000 (30 min) Per-phase timeout in ms
--keep-alive off Keep sandboxes running after eval
--keep-hours N 8 Hours to keep alive (with --keep-alive)
--skip-verify off Skip the agent-browser verification phase
--skip-deploy off Skip the Vercel deploy phase
--scenarios a,b,c all Only run specific scenarios by slug
--scenarios-file path Load scenarios from a JSON file instead of built-in defaults

Read the full file on GitHub · 388 lines

Files

What ships with it

8 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 10d ago First seen · 388 lines · 51 tokens per session scan B 2588bbda779e

Subscribe to this mod's changes

benchmark-sandbox is a skill published in the GitHub repository build-with-dhiraj/ai-workflow-framework-portability-kit (4 stars, last pushed 28d ago), licensed MIT. It adds 51 tokens to every session and 5,749 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 2 findings (reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

semgrep-rule-variant-creator

Creates language variants of existing Semgrep rules. Use when porting a Semgrep rule to specified target languages. Takes an existing rule and target languages as input, produces independent rule+test directories for each language.

waybarrios/opencode-power-pack · 50 tokens

forward-derivation

Derive BDD scenarios and TDD test skeletons from approved SDD specifications. ATDD acceptance test tables are optional output for specialized needs. Use when: spec is approved, starting BDD/TDD implementation, generating test structures. Keywords: forward derivation, spec to test, BDD generation, TDD skeleton, test…

ValorVie/custom-skills · 91 tokens

atdd-assistant

A guide to Acceptance Test-Driven Development, or ATDD, a process where a team agrees on user-story requirements and acceptance tests before coding. Acceptance tests check whether a feature meets the agreed business requirements.

ValorVie/custom-skills · 68 tokens

testing-guide

A testing guide covering common testing levels, including unit, integration, system, and end-to-end testing. It also supports ISTQB and industry testing-pyramid approaches.

ValorVie/custom-skills · 81 tokens

bdd-assistant

A guide to Behavior-Driven Development, or BDD, a way to describe software behavior with examples in plain language before building it. It uses Given-When-Then scenarios and Gherkin, a structured format for those examples.

ValorVie/custom-skills · 76 tokens

methodology-system

Manage and guide developers through active development methodology workflows. Use when: TDD, BDD, SDD, ATDD, or custom methodology workflows are needed. Keywords: methodology, workflow, TDD, BDD, SDD, ATDD, phase, checkpoint, development process.

ValorVie/custom-skills · 61 tokens