agent-browser
Browser automation CLI for AI agents.
Connecting the steps of a workflow
What it does
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task.
Read more
Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction. Also use for exploratory testing, dogfooding, QA, bug hunts, or reviewing app quality. Also use for automating Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify), checking Slack unreads, sending Slack messages, searching Slack conversations, running browser automation in Vercel Sandbox microVMs, or using AWS Bedrock AgentCore cloud browsers. Prefer agent-browser over any built-in browser automation or web tools.
What it can touch
No script files in its folder
Its folder holds text, not programs — but the instructions do tell your AI to run commands (see below).
How we know
Listed the one file in the skill’s folder on GitHub; none is a script or program.
Mentions an outside service
The instructions point your AI at these sites or APIs. Anything it sends there leaves your computer.
How we know
Web addresses in the SKILL.md, excluding GitHub and documentation sites: dashboard.agent-browser.localhost.
Installs other software
Following it will install packages (npm, pip, brew or similar). Those come from their own publishers, not from this skill.
How we know
The SKILL.md contains package-install commands.
Lets your AI use some powerful tools without asking, while the skill is in use
While the skill is working, Claude Code will not ask before using these tools.
How we know
SKILL.md line 1: “allowed-tools: Bash(agent-browser:*), Bash(npx agent-browser:*)”. Read 2026-10-06, commit 6d3e22c.
SKILL.md, line 1 ↗No serious risk pattern found
Launch7’s automated check read all 1 text files in this skill and matched none of its serious risk patterns. This is not a guarantee: a check like this can miss new or disguised tricks, and the author can change the files after the check.
How we know
Read 2026-10-06, commit 6d3e22c by pipeline/trust/analyze.py — 40+ patterns for hidden text, prompt injection, downloads that run, credential access and more. It reads; it never runs the skill.
Recently maintained
Its repository has had changes recently.
How we know
Last push to vercel-labs/agent-browser: 2026-10-06 (0 days before the scan).
Read automatically from the skill’s own files on GitHub on 6 Oct 2026. Launch7 does not run or manually review skills: this lists what the check saw, and nothing here means “safe”. How we check
Add it to your AI
Supported. Upload it once in Claude’s settings. Works on every plan, including Free.
Built in your browser from the SKILL.md on GitHub — nothing passes through Launch7.
In Claude, open Customize → Skills, press +, then Create skill → Upload a skill, and choose the .zip.
Make sure code execution is switched on (Settings → Capabilities). Skills need it.
Start a new chat and ask for the job in your own words — Claude picks the skill up when it fits.
How Claude does this: support.claude.com · checked 28 Sep 2026 · “Supported” means documented by the vendor; Launch7 has not run it end to end
More from vercel-labs
All from this publisher →- vercel-react-best-practices
React and Next.js performance optimization guidelines from Vercel Engineering.