Atlas / MCP servers / angiejones / Selenium

SeleniumSAFE

mcp/angiejones/selenium

An MCP implementation for Selenium WebDriver

Verdict
SAFE
Grade
B
Trust score
89 /100
Exposed tools
18 12r · 5w · 1d
Transport
stdio
License
MIT
Stars
426
01

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

A Model Context Protocol (MCP) server for Selenium WebDriver — browser automation for AI agents.

[](https://youtu.be/mRV0N8hcgYA)

Setup

Goose (Desktop)

Paste into your browser address bar:

goose://extension?cmd=npx&arg=-y&arg=%40angiejones%2Fmcp-selenium%40latest&id=selenium-mcp&name=Selenium%20MCP&description=automates%20browser%20interactions

Goose (CLI)

goose session --with-extension "npx -y @angiejones/mcp-selenium@latest"

Claude Code

claude mcp add selenium -- npx -y @angiejones/mcp-selenium@latest

Cursor / Windsurf / other MCP clients

{
"mcpServers": {
"selenium": {
"command": "npx",
"args": ["-y", "@angiejones/mcp-selenium@latest"]
}
}
}

Example Usage

Tell the AI agent of your choice:

Open Chrome, go to github.com/angiejones, and take a screenshot.

The agent will call Selenium's APIs to start_browser, navigate, and take_screenshot. No manual scripting or explicit directions needed.

Supported Browsers

Chrome, Firefox, Edge, and Safari.

Safari note: Requires macOS. Run sudo safaridriver --enable once and enable "Allow Remote Automation" in Safari → Settings → Developer. No headless mode.

Tools

start_browser

Launches a browser session.

Read from source at commit a1fb02657c08OBSERVED · 2026-10-01
02

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control.

claude-code
claude mcp add mcp-selenium -- npx -y @angiejones/[email protected]
claude-desktop
{
  "mcpServers": {
    "mcp-selenium": {
      "command": "npx",
      "args": [
        "-y",
        "@angiejones/[email protected]"
      ]
    }
  }
}
03

Exposed tools (18)

12 read · 5 write · 1 destructive. Blast radius: 1 tool can delete or overwrite — an agent that can be talked into calling a tool can be talked into calling this one.

ToolRiskDescription
add_cookiewrite
alertread
close_sessionread
delete_cookiedestructive
diagnosticsread
execute_scriptwrite
frameread
get_cookiesread
get_element_attributeread
get_element_textread
interactread
navigateread
press_keyread
send_keyswrite
start_browserwrite
take_screenshotread
upload_filewrite
windowread
04

Trust audit

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

LayerWhat it checksResult
L0Provenance & inventoryPASS
L1Static analysis of the codePASS
L2Instruction surface (what it tells the agent)PASS
L3Class-specific surfaceWARN
L4Behavioural (sandbox)SKIPPED

What the source does

Filesystem
declared (2 observation(s))
Network
none-observed
Shell
none-observed
Dependencies
not all pinned
Secrets in source
none-found

Findings (4)

MEDIUMFilesystem / path · mcp.destructive_tools · CWE-22, CWE-59
delete_cookie
Why it matters. 1 tool(s) can delete or overwrite
Fix. prefer a read-only mode or scoped tokens; the page states the blast radius
LOWFilesystem / path · fs.traversal · CWE-22, CWE-59
src/lib/server.js:18
const { version } = require('../../package.json');
LOWNetwork egress · net.raw_ip · CWE-200, CWE-319
test/cookies.test.mjs:28
resolve({ server, url: `http://127.0.0.1:${port}` });
LOWSupply chain · supply.unpinned · CWE-829, CWE-1357
package.json
@modelcontextprotocol/sdk, selenium-webdriver
Why it matters. 2 dependency range(s) float
Fix. pin exact versions or ship a lockfile

Gates applied: no_behavioural_pass.

Audited 2026-10-01 · audit v0.4.1 · source sha a1fb02657c08full audit observations/trust-audit/mcp-server/angiejones__selenium.json · Report an issue / request a re-scan
05

Audit history

Every audit this server has had. A grade with a past is a grade somebody is still checking.

DateSourceVerdictGradeScoreChange
2026-10-01a1fb02657c08SAFEB89first audit
06

Questions

What is the Selenium MCP server?

An MCP implementation for Selenium WebDriver

What tools does Selenium expose?

18 in total: 12 read-only, 5 that write, and 1 that can delete or overwrite (delete_cookie). Every one is listed on this page with its risk.

Is Selenium safe to connect to an agent?

The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B. Separately from the audit: 1 of its tools can destroy data, so scope the token you give it to what you actually need.

What credentials does Selenium need?

No credential environment variables were found in its source, so it appears to need none.

How does Selenium run?

It speaks stdio, so it runs as a local process your client starts. It is published on npm as @angiejones/mcp-selenium at 0.2.3.

How current is this page?

The grade is for one exact copy of the source (a1fb02657c08), read on 2026-10-01. The repository is watched and re-audited when it changes.

Advertisement