Skills/npm/real-browser-mcp-server

real-browser-mcp-server@3.3.17

VERIFIED

Publisher claims: "MCP Server for Real Browser - Patchright Blocker." Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry npm (MCP server)Publisher codeiva · publishing for 178 daysInstalls 2,921Last detonated 3d ago
$claude mcp add real-browser-mcp-server -- npx -y real-browser-mcp-server
WatchReport an issue

Coverage

Server started and answered tools/list.
Tested
install scripts, with decoy credentials in place · second launch with the clock 45 days ahead · MCP initialize and tools/list · every tool description, for hidden instructions · every file read, file written and host contacted by what ran
Not tested
what the tools do when called with real credentials

Before you give it your token

It will ask for
no credential variables read in the code
Where it talks to
no vendor host named in the code
What the honeypot saw
no credential files read · no network
What we could not test
what the tools do when called with real credentials

Reasonable to connect. No credential asked for, no third-party host named, Server started and answered tools/list.

Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (3 scenarios) on 2026-11-11 and scored it 92/100 (Verified). It did what it claims and nothing else we could detect: no undeclared network egress, no credential reads, no persistent instruction writes. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 27 s · 3 scenarios · 0 deviations
t+CategoryObserved action
0.0sprocessexited 127: bash: line 1: patchright: command not found
6.6sprocessdependency install exit 0
27sclaimed19 tools: browser_init, navigate, get_content, wait, click, type, browser_close, solve_captcha, random_scroll, redirect_tracer, extract_data, press_key, deep_analysis, network_recorder, media_extractor, execute_js, storage_inspector, see_page, async_job
27sprocessscript backend runs setup only
27sprocesssecond launch, clock +45 days, CI unset
27sprocessexited 127: bash: line 1: patchright: command not found

Static findings

MEDIUM
Process kill commanddist/src/shared/tools.js:205 · Process kill command — agent can terminate processesforce: { type: 'boolean', default: false, description: 'Force-kill the browser process' }Fix: Build commands from fixed arguments; do not interpolate untrusted strings into a shell.
AST05

No other findings across 11,670 lines in 72 files.

Cost to the agent

Instructions loaded
≈54 tok
1 file · 72 in bundle
Tool list per prompt
≈7,097 tok
19 tools · 374 per tool
Handshake
662 ms
launch to tools/list, stdio
Undocumented tools
0
description under 20 chars · avg 211

Tokens estimated at four characters each. The tool list is re-read by the agent on every turn, so it is the number to compare between servers that do the same job: ranked across the index.

Known vulnerabilities

No advisories in OSV for real-browser-mcp-server@3.3.17 as of 3d ago. This is the package itself, not its dependencies.