CtrlK
BlogDocsLog inGet started
Tessl Logo

vercel-sandbox

Run agent-browser + Chrome inside Vercel Sandbox microVMs for browser automation from any Vercel-deployed app. Use when the user needs browser automation in a Vercel app (Next.js, SvelteKit, Nuxt, Remix, Astro, etc.), wants to run headless Chrome without binary size limits, needs persistent browser sessions across commands, or wants ephemeral isolated browser environments. Triggers include "Vercel Sandbox browser", "microVM Chrome", "agent-browser in sandbox", "browser automation on Vercel", or any task requiring Chrome in a Vercel Sandbox.

74

Quality

91%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

82%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.

An actionable, well-organized guide dominated by executable TypeScript examples and useful tables. Minor conciseness gains are possible by deduplicating the screenshot pattern and trimming the analogy.

Suggestions

Factor the shared screenshot→base64→close block into one example and reference it, instead of repeating it in the Screenshot and Multi-Step Workflows sections.

Add a brief validate→fix→retry note for the form-submission workflow (e.g. re-snapshot after wait to confirm the expected state) to strengthen feedback loops.

Consider moving the cron and framework-location material into a separate reference file to keep SKILL.md as a lean overview, since the body exceeds 50 lines.

DimensionReasoningScore

Conciseness

Mostly lean code-first content, but the Docker-image analogy and the repeated base64-screenshot pattern across two examples could be trimmed for a fully token-efficient body.

4 / 5

Actionability

Provides complete, copy-paste-ready TypeScript functions covering screenshot, accessibility snapshot, multi-step form fill, and cron cases, with specific commands and env-var/framework tables.

5 / 5

Workflow Clarity

Sequences multi-step workflows clearly (open → snapshot → fill/click → wait → screenshot → close) with one explicit checkpoint ('if (!ssPath) throw'), but lacks a full validate→fix→retry feedback loop.

4 / 5

Progressive Disclosure

Well-organized into clearly-headed sections in a single file with one clearly-signaled external reference (examples/environments/); no bundle files exist to split content further.

4 / 5

Total

17

/

20

Passed

Description

100%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 strong, well-structured description that clearly conveys both capability and trigger conditions with concrete, natural-language terms. Third-person voice is maintained throughout.

DimensionReasoningScore

Specificity

Lists multiple concrete capabilities — 'run headless Chrome without binary size limits', 'persistent browser sessions across commands', 'ephemeral isolated browser environments' — giving comprehensive coverage of the niche.

5 / 5

Completeness

Explicitly states what it does ('Run agent-browser + Chrome inside Vercel Sandbox microVMs') and when to use it via a 'Use when...' clause with concrete trigger phrases.

5 / 5

Trigger Term Quality

Includes natural phrases users would say ('Vercel Sandbox browser', 'microVM Chrome', 'agent-browser in sandbox', 'browser automation on Vercel') plus framework names and synonyms.

5 / 5

Distinctiveness Conflict Risk

Tightly scoped to Vercel Sandbox microVMs + Chrome browser automation with distinct triggers, giving minimal conflict risk with other skills.

5 / 5

Total

20

/

20

Passed

Validation

100%

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
vercel-labs/agent-browser
Reviewed

Table of Contents

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.