Brave Real Browser MCP Server
brave-real-browser-mcp-server
npm
v2.45.29
Published by codeiva4u — no publish provenance, so origin is unverified, but the source is public: the repository link below is self-declared yet readable, so you can inspect the code before adopting it.
MCP Server for Brave Real Browser - Puppeteer with Brave Browser, Stealth Mode, Ad Blocker, and Turnstile Auto-Solver for undetectable web automation.
The grade answers one question — how safe is this server for you to adopt — so it is computed in two auditable stages. Nothing below is an opinion or an LLM's guess; every line is a real term the deterministic engine applied, and the same input always yields the same number.
1. Threat score — 100 − 14.1 = 86. What the published surface and source actually contain:
| Points | What was found | Category |
|---|---|---|
| −12.9 | Untrusted input concatenated into a command sink ×6 MTC-SRC-009 | injection |
| −1.2 | Package runs install-time scripts MTC-SUP-010 | supply-chain |
2. Client adoption risk — 86 − 7 = 79. Three small, subtract-only factors that reflect your risk in adopting it — a clean scan proves less on a powerful, unverified or barely-inspectable package, so the grade says so plainly:
| Points | Adoption-risk factor |
|---|---|
| −6 | capability blast radius (high) — client exposure if the model is manipulated |
| −1 | publisher verification (public source) — no provenance, but the source is public and inspectable |
Capability observations and info notes are shown under Findings but never scored.
Open any row's finding below for the file, line and evidence behind a deduction.
Findings 25
high Dynamic code execution in server code (packages/brave-real-playwright-core/lib/generated/injectedScriptSource.js)MTC-SRC-001
In the server's implementation (`packages/brave-real-playwright-core/lib/generated/injectedScriptSource.js:24`): Evaluating strings as code is the most direct RCE primitive; if any tool input reaches it, the server executes attacker-chosen code. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: dScript = this;\n }\n eval(expression) {\n return this.window.eval(expression);\n }\n testIdAttributeNameForStric
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-playwright-core/lib/generated/injectedScriptSource.js
high Dynamic code execution in server code (packages/brave-real-playwright-core/lib/server/javascript.js)MTC-SRC-001
In the server's implementation (`packages/brave-real-playwright-core/lib/server/javascript.js:236`): Evaluating strings as code is the most direct RCE primitive; if any tool input reaches it, the server executes attacker-chosen code. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: tion) { try { new Function("(" + expression + ")"); } catch (e1) { if (expression.startsWith("async
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-playwright-core/lib/server/javascript.js
high Dynamic code execution in server code (packages/brave-real-playwright-core/lib/vite/traceViewer/assets/defaultSettingsView-BEpdCv1S.js)MTC-SRC-001
In the server's implementation (`packages/brave-real-playwright-core/lib/vite/traceViewer/assets/defaultSettingsView-BEpdCv1S.js:116`): Evaluating strings as code is the most direct RCE primitive; if any tool input reaches it, the server executes attacker-chosen code. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: .__injectedScript=this)}eval(e){return this.window.eval(e)}testIdAttributeNameForStrictErrorAndConsoleCodegen(){return t
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-playwright-core/lib/vite/traceViewer/assets/defaultSettingsView-BEpdCv1S.js
high Dynamic code execution in server code (src/lsp/capabilities/diagnostics.js)MTC-SRC-001
In the server's implementation (`src/lsp/capabilities/diagnostics.js:299`): Evaluating strings as code is the most direct RCE primitive; if any tool input reaches it, the server executes attacker-chosen code. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: (') || jsCode.includes('Function(')) { diagnostics.push({ severity: DiagnosticSeverity.Warning,
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server src/lsp/capabilities/diagnostics.js
high Shell/command execution in server code (packages/brave-real-launcher/src/brave-finder.ts)MTC-SRC-002
In the server's implementation (`packages/brave-real-launcher/src/brave-finder.ts:11`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: port { execSync } from 'child_process'; import escapeRegExp from 'escape-string-regexp'; import log from './logger.js';
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/src/brave-finder.ts
high Shell/command execution in server code (packages/brave-real-launcher/src/brave-installer.ts)MTC-SRC-002
In the server's implementation (`packages/brave-real-launcher/src/brave-installer.ts:14`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: execSync, spawn } from 'child_process'; import { homedir } from 'os'; import log from './logger.js'; import { getPlatfor
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/src/brave-installer.ts
high Shell/command execution in server code (packages/brave-real-launcher/src/brave-launcher.ts)MTC-SRC-002
In the server's implementation (`packages/brave-real-launcher/src/brave-launcher.ts:8`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: * as childProcess from 'child_process'; import * as fs from 'fs'; import * as net from 'net'; import * as path from 'pat
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/src/brave-launcher.ts
high Shell/command execution in server code (packages/brave-real-launcher/src/chrome-launcher.ts)MTC-SRC-002
In the server's implementation (`packages/brave-real-launcher/src/chrome-launcher.ts:8`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: * as childProcess from 'child_process'; import * as fs from 'fs'; import * as net from 'net'; import * as path from 'pat
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/src/chrome-launcher.ts
high Shell/command execution in server code (packages/brave-real-launcher/src/utils.ts)MTC-SRC-002
In the server's implementation (`packages/brave-real-launcher/src/utils.ts:9`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: port childProcess from 'child_process'; import { mkdirSync } from 'fs'; import isWsl from 'is-wsl'; import which from 'w
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/src/utils.ts
high Shell/command execution in server code (packages/brave-real-launcher/src/xvfb-support.ts)MTC-SRC-002
In the server's implementation (`packages/brave-real-launcher/src/xvfb-support.ts:8`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: * as childProcess from 'child_process'; import * as fs from 'fs'; import {getPlatform} from './utils.js'; import log fro
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/src/xvfb-support.ts
medium Hardcoded egress to an external endpoint (packages/brave-real-launcher/src/chrome-version.ts)MTC-SRC-003
In the server's implementation (`packages/brave-real-launcher/src/chrome-version.ts:33`): A hardcoded outbound call to a fixed external host inside server code is a classic exfiltration/telemetry channel — especially paired with reads of local data. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: const response = await fetch( 'https://chromiumdash.appspot.com/fetch_releases?channel=Stable&platform=Wind
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/src/chrome-version.ts
medium Untrusted input concatenated into a command sink (packages/brave-real-launcher/src/brave-finder.ts)MTC-SRC-009
In the server's implementation (`packages/brave-real-launcher/src/brave-finder.ts:64`): A shell/process command assembled from concatenated or interpolated values is command injection when any part is attacker-influenced — the OWASP canonical RCE flow. Verify what reaches the interpolated value. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: stallations try { execSync( `${LSREGISTER} -dump` + ' | grep -i \'brave browser\\( beta\\| nightly\\|
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/src/brave-finder.ts
medium Untrusted input concatenated into a command sink (packages/brave-real-launcher/src/brave-installer.ts)MTC-SRC-009
In the server's implementation (`packages/brave-real-launcher/src/brave-installer.ts:105`): A shell/process command assembled from concatenated or interpolated values is command injection when any part is attacker-influenced — the OWASP canonical RCE flow. Verify what reaches the interpolated value. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: try { execSync(`"${installerPath}" /silent /install`, { stdio: 'ignore',
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/src/brave-installer.ts
medium Untrusted input concatenated into a command sink (packages/brave-real-launcher/src/brave-launcher.ts)MTC-SRC-009
In the server's implementation (`packages/brave-real-launcher/src/brave-launcher.ts:770`): A shell/process command assembled from concatenated or interpolated values is command injection when any part is attacker-influenced — the OWASP canonical RCE flow. Verify what reaches the interpolated value. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: const taskkillProc = spawnSync( `taskkill /pid ${this.braveProcess.pid} /T /F`, { shell: true, encoding: 'u
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/src/brave-launcher.ts
medium Untrusted input concatenated into a command sink (packages/brave-real-launcher/src/chrome-launcher.ts)MTC-SRC-009
In the server's implementation (`packages/brave-real-launcher/src/chrome-launcher.ts:546`): A shell/process command assembled from concatenated or interpolated values is command injection when any part is attacker-influenced — the OWASP canonical RCE flow. Verify what reaches the interpolated value. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: const taskkillProc = spawnSync( `taskkill /pid ${this.chromeProcess.pid} /T /F`, { shell: true, encoding: '
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/src/chrome-launcher.ts
medium Untrusted input concatenated into a command sink (packages/brave-real-launcher/src/utils.ts)MTC-SRC-009
In the server's implementation (`packages/brave-real-launcher/src/utils.ts:183`): A shell/process command assembled from concatenated or interpolated values is command injection when any part is attacker-influenced — the OWASP canonical RCE flow. Verify what reaches the interpolated value. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: { childProcess.execSync(`xdpyinfo -display ${this.display}`, { stdio: 'ignore', timeout: 10
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/src/utils.ts
medium Untrusted input concatenated into a command sink (packages/brave-real-playwright-core/lib/server/utils/processLauncher.js)MTC-SRC-009
In the server's implementation (`packages/brave-real-playwright-core/lib/server/utils/processLauncher.js:194`): A shell/process command assembled from concatenated or interpolated values is command injection when any part is attacker-influenced — the OWASP canonical RCE flow. Verify what reaches the interpolated value. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: lProcess = childProcess.spawnSync(`taskkill /pid ${spawnedProcess.pid} /T /F`, { shell: true }); const [stdout
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-playwright-core/lib/server/utils/processLauncher.js
low Hardcoded egress to an external endpoint in packaging/dev tooling (packages/brave-real-puppeteer-core/scripts/chrome-version.js)MTC-SRC-003
In a packaging/dev/install script (shipped, but not the server runtime) (`packages/brave-real-puppeteer-core/scripts/chrome-version.js:23`): A hardcoded outbound call to a fixed external host inside server code is a classic exfiltration/telemetry channel — especially paired with reads of local data. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: const response = await fetch( 'https://chromiumdash.appspot.com/fetch_releases?channel=Stable&platform=Wind
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-puppeteer-core/scripts/chrome-version.js
low Hardcoded egress to an external endpoint in packaging/dev tooling (packages/brave-real-puppeteer-core/scripts/proxy-manager.js)MTC-SRC-003
In a packaging/dev/install script (shipped, but not the server runtime) (`packages/brave-real-puppeteer-core/scripts/proxy-manager.js:164`): A hardcoded outbound call to a fixed external host inside server code is a classic exfiltration/telemetry channel — especially paired with reads of local data. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: const response = await fetch('https://httpbin.org/ip', { signal: controller.signal, // Note: In
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-puppeteer-core/scripts/proxy-manager.js
low Shell/command execution in packaging/dev tooling (packages/brave-real-launcher/scripts/chrome-launcher-sync.cjs)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`packages/brave-real-launcher/scripts/chrome-launcher-sync.cjs:11`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: { execSync } = require('child_process'); class ChromeLauncherSyncManager { constructor() { this.projectRoot = pro
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/scripts/chrome-launcher-sync.cjs
low Shell/command execution in packaging/dev tooling (packages/brave-real-launcher/scripts/postinstall.cjs)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`packages/brave-real-launcher/scripts/postinstall.cjs:12`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: { execSync } = require('child_process'); const path = require('path'); const fs = require('fs'); const colors = { g
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/scripts/postinstall.cjs
low Shell/command execution in packaging/dev tooling (packages/brave-real-launcher/scripts/sync-chrome-launcher.cjs)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`packages/brave-real-launcher/scripts/sync-chrome-launcher.cjs:11`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: { execSync } = require('child_process'); class ChromeLauncherSync { constructor() { this.chromeRepo = 'GoogleChro
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/scripts/sync-chrome-launcher.cjs
low Shell/command execution in packaging/dev tooling (packages/brave-real-launcher/scripts/version-increment.cjs)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`packages/brave-real-launcher/scripts/version-increment.cjs:11`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: { execSync } = require('child_process'); class VersionManager { constructor() { this.projectRoot = process.cwd();
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/scripts/version-increment.cjs
low Dynamic module load from a non-literal in packaging/dev tooling (packages/brave-real-launcher/scripts/postinstall.cjs)MTC-SRC-005
In a packaging/dev/install script (shipped, but not the server runtime) (`packages/brave-real-launcher/scripts/postinstall.cjs:46`): Loading a module chosen at runtime (from a variable) can pull in and run attacker-influenced code paths. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: const module = await import(fileUrl); const { getInstallations, BraveInstaller } = module; // Check
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server packages/brave-real-launcher/scripts/postinstall.cjs
low Package runs install-time scripts (postinstall)MTC-SUP-010
"brave-real-browser-mcp-server" executes postinstall script(s) at install time. An install hook runs at install time; most are routine build/setup, but review what it does before trusting it.
Evidence: npm run build:all && node scripts/auto-update-deps.js || true
Fix: Review the scripts; install with --ignore-scripts where possible and vet what they do.
Location: package brave-real-browser-mcp-server
Tools 1
Each tool and what it can reach — statically extracted from the published source.
brave-real-browser-mcp-serverno sensitive capability
What this scan could not see
- Tools were statically extracted from the published source (1 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
Versions 1
Scan history per published version. The engine is deterministic — the same version always yields the same score, so a changed score means the package itself changed.
| Version | Score | Findings | Engine | Scanned |
|---|---|---|---|---|
v2.45.29 latest |
C 79/100 | 25 | 1.13.0 | 2026-09-07 |
Embed this score
Show this server's live Trust Score in your README, docs or website. The badge is served straight from the registry and updates automatically after every rescan — no API key needed. It links back to this page, so anyone who sees the grade can also read the findings behind it instead of taking a number on faith.
Verify this score yourself
The score above is reproducible: the same package version always yields the same result. Run it locally or over the free API — no account, no LLM, fully deterministic.
npx mcptrustchecker scan brave-real-browser-mcp-server --onlineMore in Browser Automation
Model Context Protocol (MCP) server that integrates AgentQL data extraction capabilities.
Screenshot any URL or HTML as PNG/JPEG/WebP from your AI agent. Full-page, clean, no install.
MCP server for aria51 accessibility scanner
Bridge any browser web app to Claude Code via MCP
Browserbase’s official MCP server: cloud headless browsers for agents, with sessions and screenshots.
MCP server for AI web browser automation using Browserbase and Stagehand