vercel-sandbox

Vercel Sandbox guidance — ephemeral Firecracker microVMs for running untrusted code safely. Supports AI agents, code generation, and experimentation. Use when executing user-generated or AI-generated code in isolation.

Quality

73%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./plugins/vercel/skills/vercel-sandbox/SKILL.md

Quality

Content

76%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

Highly actionable content with comprehensive, executable code examples and clear section structure. The main gaps are missing validation checkpoints in the workflows and minor redundancy from the repeated dependency list.

Suggestions

Add explicit verification steps (e.g., check the screenshot path is non-empty, assert page title) and a validate/retry feedback loop around form submission.

Extract the duplicated CHROMIUM_SYSTEM_DEPS array into a single shared snippet or reference to remove redundancy.

Consider moving the snapshot-creation walkthrough into a separate reference file to keep SKILL.md a leaner overview.

Dimension	Reasoning	Score
Conciseness	Mostly lean and code-heavy with minimal prose, but the CHROMIUM_SYSTEM_DEPS array is duplicated across two sections and the Docker analogy adds slight over-explanation.	4 / 5
Actionability	Fully executable, copy-paste-ready TypeScript covering the common cases (screenshot, accessibility snapshot, multi-step forms, cron) plus concrete install/env commands.	5 / 5
Workflow Clarity	Sequences are clear via the withBrowser helper, but there are no explicit validation/verification checkpoints, and outward-facing operations like form submission lack error-recovery feedback loops.	3 / 5
Progressive Disclosure	Well-organized into clearly headed sections with a signaled external pointer to examples/environments/; minor gaps include the inlined duplicated snapshot code that could be factored out.	4 / 5
	Total	16 / 20 Passed

Description

70%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A solid description that states a clear niche (Vercel Sandbox microVMs) and provides an explicit 'Use when...' trigger. The main weakness is that the 'what' leans on the vague word 'guidance' and the listed actions are somewhat generic.

Suggestions

Replace 'guidance' with concrete verbs (e.g., 'Runs untrusted code inside ephemeral Firecracker microVMs') to sharpen the 'what'.

Add natural trigger synonyms users say ('isolated environment', 'sandboxed execution') to broaden keyword coverage.

Tie the 'when' clause more specifically to the browser-automation use case the skill actually targets.

Dimension	Reasoning	Score
Specificity	Names the domain ('ephemeral Firecracker microVMs for running untrusted code safely') and several actions ('AI agents, code generation, and experimentation'), but the actions are high-level and 'guidance' is the operative noun, so coverage is not comprehensive.	3 / 5
Completeness	Both 'what' (ephemeral microVMs for running untrusted code) and 'when' ('Use when executing user-generated or AI-generated code in isolation') are present and explicit, but the 'what' is slightly softened by the word 'guidance'.	4 / 5
Trigger Term Quality	Good natural keyword coverage ('Sandbox', 'untrusted code', 'code in isolation', 'AI-generated code'), though a few common variations like 'isolated environment' or 'sandboxed execution' are absent.	4 / 5
Distinctiveness Conflict Risk	Tied to a specific product (Vercel Sandbox) and use case with distinct triggers, leaving only minor overlap risk with generic sandboxing/isolation skills.	4 / 5
	Total	15 / 20 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	14 / 16 Passed

Repository: openai/plugins
Path: plugins/vercel/skills/vercel-sandbox/SKILL.md
Commit: 1e28582

Reviewed: 6 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.