# Agent Browser

> Use this tool when you need to automate browser interactions, such as navigation, form filling, and screenshot capture, with a range of 44 automation tools. It solves problems related to web scraping, testing, and automation, providing a token-efficient referencing system and supporting both headless and headed mode. It takes page snapshots and other inputs, and outputs screenshots, PDFs, and video recordings, making it ideal for use cases requiring efficient and flexible browser automation.

Canonical page: https://skillsregistry.net/skills/hughedward-agent-browser  
JSON: https://api.skillsregistry.net/v1/skills/hughedward-agent-browser

## Description

MCP server wrapping Vercel Labs' agent-browser CLI tool, providing 44 browser automation tools through Playwright. Features a token-efficient @ref element referencing system that reduces context usage by generating stable references from page snapshots. Supports navigation, form filling, clicking, scrolling, screenshot capture, PDF export, tab management, iframe switching, dialog handling, drag-and-drop, network interception, video recording, and session state persistence. Launches browsers on demand with both headless and headed mode support.

## Trust

- **Trust score (0–1):** 0.63
- **Verification tier:** scanned
- **Last scanned:** 2026-09-02

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** browser-automation
- **Updated:** 2026-09-02

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/hughedward-agent-browser)
- **Repository:** <https://github.com/hughedward/agent_browser_mcp>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "hughedward-agent-browser"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/hughedward-agent-browser` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/hughedward-agent-browser/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt