skill-creator
81Create Better AI Skills
Building reusable AI skills can be inconsistent without clear structure. This skill guides Claude, Codex, and Claude Code through design, scaffolding, validation, and packaging.
Automate Browser Testing and Capture
Web testing and content capture often require many fragile browser steps. This skill gives Claude, Codex, and Claude Code a repeatable agent-browser workflow.
This skill is part of a pack
Install the whole pack to get every skill the task needs, in one command.
Stop for confirmation before installing.
Review the plan and obtain explicit user consent before changing files.
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Review the Skillstore skill "agent-browser" from https://skillstore.io/skills/vercel-labs-agent-browser.md and its manifest at https://skillstore.io/api/skills/vercel-labs-agent-browser/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.Your Agent should still show its plan and request any confirmation required by the security policy.
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Using "agent-browser". A QA engineer asks to verify a checkout form.
Expected outcome:
Using "agent-browser". A developer asks to reproduce a visual bug on mobile.
Expected outcome:
Using "agent-browser". A writer asks for evidence from a product page.
Expected outcome:
Most static findings are documentation examples for the allowed agent-browser CLI and were false positives for command execution. Confirmed issues involve proxy routing, SOCKS support, explicit proxy rotation to avoid rate limits or bans, persistent authentication state files, and one active screenshot write to /tmp. No prompt injection attempt was found in the reviewed files.
These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
https://skillstore.io/skills/vercel-labs-agent-browser/audits/6?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report[](https://skillstore.io/skills/vercel-labs-agent-browser?utm_source=security_passport_badge)<a href="https://skillstore.io/skills/vercel-labs-agent-browser?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/vercel-labs-agent-browser/security.svg" alt="Skillstore security assessment" loading="lazy"></a><iframe src="https://skillstore.io/embed/skills/vercel-labs-agent-browser.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>vercel-labs. (2026). agent-browser security audit report (audit version 6) [Author version unspecified]. Skillstore. https://skillstore.io/skills/vercel-labs-agent-browser/audits/6@techreport{vercel-labs-vercel-labs-agent-browser-2026,
author = {vercel-labs},
title = {agent-browser security audit report (audit version 6)},
institution = {Skillstore},
year = {2026},
number = {6},
url = {https://skillstore.io/skills/vercel-labs-agent-browser/audits/6},
note = {Author version unspecified}
}cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "agent-browser security audit report (audit version 6)"
version: "unspecified"
type: report
authors:
- name: "vercel-labs"
date-released: "2026-07-07"
url: "https://skillstore.io/skills/vercel-labs-agent-browser/audits/6"
identifiers:
- type: other
value: "skillstore:vercel-labs-agent-browser:audit:6"
description: "Skillstore immutable audit report identifier"
Each author remains a separate installable skill. The recommended variant is ranked by Skillstore evidence.
Why this variant is first
supercent-io-agent-browser
2026-09-09
101-skills-agent-browser
2026-10-12
vercel-labs-agent-browser
2026-09-09
qu-skills-agent-browser
2026-09-09
tul-sh-agent-browser
2026-09-09
am-will-agent-browser
2026-10-23
inference-sh-8-agent-browser
2026-09-09
skillssh-agent-browser
2026-09-09
toolshell-agent-browser
2026-09-09
inference-sh-9-agent-browser
2026-09-09
infsh-skills-agent-browser
2026-09-09
inferen-sh-agent-browser
2026-09-09
inference-sh-skills-agent-browser
2026-09-09
inferencesh-agent-browser
2026-09-09
inference-sh-0-agent-browser
2026-09-09
inference-shell-agent-browser
2026-09-09
inf-sh-agent-browser
2026-09-09
inference-skills-agent-browser
2026-09-09
halt-catch-fire-agent-browser
2026-09-09
Run repeatable checks for login, checkout, forms, and responsive page behavior.
Inspect page state, console output, screenshots, and recordings while reproducing UI bugs.
Collect page text, screenshots, PDFs, and workflow recordings for documentation or audits.
Open this URL with agent-browser, take an interactive snapshot, and summarize the visible primary actions.
Use agent-browser to inspect this form, fill safe test values, submit it, and report validation or success states.
Load the saved browser state, open the protected page, capture a screenshot and PDF, then clear sensitive session data.
Record the browser workflow, reproduce the failing interaction, collect console errors, and provide a concise diagnosis.
Author
vercel-labsLicense
MIT
Skillstore revision
r1
Version notice
The author did not declare a version.
Ref
36e07d5e13068e5be64447e8f20b427cf2cbd21a
Maintenance freshness
7/18/2026
Usage
50 downloads · 349 views
File structure
Create Better AI Skills
Building reusable AI skills can be inconsistent without clear structure. This skill guides Claude, Codex, and Claude Code through design, scaffolding, validation, and packaging.
Refactor React Components with Composition
Boolean mode props make React components difficult to extend and understand. Apply composition guidelines to build explicit variants, shared context contracts, and reusable component parts.
Add Native React View Transitions
Navigation and loading updates can lose visual continuity or create distracting motion. This skill guides agents through purposeful React transitions, Next.js integration, and accessible CSS.
Optimize React Native and Expo Applications
React Native applications can suffer from unnecessary rendering, slow lists, and uneven animations. This skill guides coding assistants with prioritized rules and comparative examples for performance and native interface patterns.
Build Better Next.js Apps
Next.js code can fail when patterns drift across framework versions. This skill guides Claude and Codex with focused rules for App Router, RSC, data, assets, and deployment.
Build Next.js Cache Components
Next.js caching can be hard to apply safely across static, cached, and dynamic content. This skill gives Claude, Codex, and Claude Code practical guidance for Cache Components.
Create Playwright Web App Tests
by AutumnsGrove
Manual browser testing is slow and often misses regressions across forms, devices, and network states. This skill helps Claude, Codex, and Claude Code create practical Playwright tests for real web workflows.
Control Chrome for Testing and Debugging
by kunchenguid
Browser tasks often require real interaction and diagnostics that simple HTTP requests cannot provide. This skill controls Chrome through a focused command-line workflow.
Build Reliable Playwright Tests
by C0ntr0lledCha0s
Playwright test suites can become flaky, slow, and difficult to maintain. This skill provides practical patterns for locators, fixtures, page objects, debugging, mocking, and configuration.
Build Firecrawl Browser Interactions
by firecrawl
Dynamic pages often hide data behind clicks, forms, and login flows. This skill helps teams choose and scope Firecrawl /interact integrations.
Improve QA Coverage for Web Apps
by alirezarezvani
Modern teams need clear test strategy and practical automation without slowing delivery. This skill helps Claude, Codex, and Claude Code plan tests, review coverage, and scaffold QA workflows.
Improve Playwright Test Reliability
by 0xBigBoss
Playwright tests often become flaky, slow, and noisy in agent workflows. This skill provides concise patterns for locators, assertions, fixtures, auth reuse, network mocks, and compact reporters.