mcp@1.47.1

VERIFIED· startup only

Publisher claims: "MCP (Model Context Protocol) server for the Browserless.io browser automation platform" Observed in sandbox: no network, touches nothing outside the project, writes no memory.

Registry npm (MCP server)Publisher browserless.io · publishing for 171 daysInstalls 18,563Last detonated 8d ago
$claude mcp add mcp -- npx -y @browserless.io/mcp
WatchReport an issue

Coverage

Startup only; authenticated tool behavior not tested.
Tested
second launch with the clock 45 days ahead · server startup over stdio (initialize sent) · every file read, file written and host contacted by what ran
Not tested
the tool list (the server exited before answering) · what the tools do when called with real credentials

What this means for the label: Verified here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.

Before you give it your token

It will ask for
AMPLITUDE_API_KEYBROWSERLESS_TOKENOAUTH_ADDITIONAL_REDIRECT_URI_PATTERNSOAUTH_ENABLEDOAUTH_REGISTER_RATE_LIMIT_PER_HOURSUPABASE_OAUTH_CLIENT_IDSUPABASE_OAUTH_CLIENT_SECRETConfiguration, not credentials: ANALYTICS_ENABLED, BROWSERLESS_API_SERVER, BROWSERLESS_API_URL, BROWSERLESS_CACHE_TTL, BROWSERLESS_MAX_RETRIES, BROWSERLESS_PUBLIC_URL, BROWSERLESS_TIMEOUT, BROWSERLESS_UPLOAD_DIRS, MCP_ALLOWED_API_URL_HOSTS, MCP_BASE_URL, MCP_COMPLIANCE_MODE, MCP_IDLE_TTL_MS, MCP_INTERNAL_PORT, MCP_SWEEP_MS, OTEL_ENABLED, OTEL_EXPORTER_OTLP_ENDPOINT, OTEL_EXPORTER_OTLP_LOGS_ENDPOINT, OTEL_METRIC_EXPORT_INTERVAL, OTEL_SDK_DISABLED, OTEL_SERVICE_NAME, REDIS_URL, SQS_QUEUE_URL, SQS_REGION, SUPABASE_SERVICE_ROLE_KEY, SUPABASE_URL, TRANSPORT
Where it talks to
no vendor host named in the codeAlso browserless.io, vscode.dev, claude.ai, chatgpt.com, grok.com, cursor.com, devin.ai, itsdev.in, make.com, celonis.com, gymshark.com — not the publisher's domain; check what for before connecting.
What we saw when it ran
no credential files read · no network
What we could not test
the tool list (the server exited before answering) · what the tools do when called with real credentials

Connect it with a scoped credential, under the guard, and read the third-party hosts first. The code names browserless.io, vscode.dev, claude.ai, chatgpt.com, grok.com, cursor.com, devin.ai, itsdev.in, make.com, celonis.com, gymshark.com as well as its own vendor. The guard injects AMPLITUDE_API_KEY, BROWSERLESS_TOKEN, OAUTH_ADDITIONAL_REDIRECT_URI_PATTERNS, OAUTH_ENABLED, OAUTH_REGISTER_RATE_LIMIT_PER_HOUR, SUPABASE_OAUTH_CLIENT_ID, SUPABASE_OAUTH_CLIENT_SECRET only into requests to the host you bind (and its subdomains) and refuses the rest.

$vet-skills run --net declared -- npx -y @browserless.io/mcp

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

What changed since 1.46.0

vet-skills diff mcp 1.46.0 1.47.1

This version reaches for something 1.46.0 did not: 11 new hosts (browserless.io, celonis.com, chatgpt.com, claude.ai, …) · 7 new credential variables (AMPLITUDE_API_KEY, BROWSERLESS_TOKEN, OAUTH_ADDITIONAL_REDIRECT_URI_PATTERNS, OAUTH_ENABLED, …). Compared from the two recorded runs, not from source.

New hosts
browserless.io, celonis.com, chatgpt.com, claude.ai, cursor.com, devin.ai, grok.com, gymshark.com, itsdev.in, make.com, vscode.dev
New credential variables
AMPLITUDE_API_KEY, BROWSERLESS_TOKEN, OAUTH_ADDITIONAL_REDIRECT_URI_PATTERNS, OAUTH_ENABLED, OAUTH_REGISTER_RATE_LIMIT_PER_HOUR, SUPABASE_OAUTH_CLIENT_ID, SUPABASE_OAUTH_CLIENT_SECRET

1.46.0 was not run in the sandbox (install failed); the comparison covers what static analysis recorded only.

The guard can hold an update like this until someone approves it (runtime.on_change: hold). A change that only shows with a real credential, or on a path the run did not reach, is not in a trace and so not in this comparison.

Is [email protected] safe to install?

KillerSkills ran it in the KillerSkills honeypot sandbox (3 scenarios; startup only; authenticated tool behavior not tested) on 2026-11-11 and scored it 84/100 (Verified). In what ran, nothing we could detect went wrong: no undeclared network egress, no credential reads, no persistent instruction writes. What it does with a real credential was not observed; the static findings and the publisher carry the rest of the score. Summary: no network, touches nothing outside the project, writes no memory.

Behavior trace

Honeypot run · 44 s · 3 scenarios · 0 deviations
t+CategoryObserved action
23sprocessdependency install exit 0
43sclaimedstartup only: exited (0) before answering · code: 'ERR_MODULE_NOT_FOUND', url: 'file:///tmp/ks-detonate-1GiQdd/task/home/work/acme-api/.claude/skills/mcp/build/src/skills/system-prompt.js' } Node.js v22.23.4
43sprocessscript backend runs setup only
43sprocesssecond launch, clock +45 days, CI unset

Static findings

HIGH
URL with embedded credentialsbuild/src/lib/agent-client.js:155 · URL with embedded credentials — may be phishing or credential leaking.describe('Bring-your-own upstream, e.g. http://user:pass@host:port'),Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST07
HIGH
Credential pass-through configurationbuild/src/lib/agent-client.js:440 · Credential pass-through configuration — agent relays sensitive credentials// logged on eviction and the URL may carry user:pass credentials. HashingFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
AST02
MEDIUM
File type combined with upload actionbuild/src/tools/agent.js:691 · File type combined with upload action — agent may exfiltrate local data filesctx.log.warn('Inline media unavailable; preserving original capture output.');Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.
AST07

Cost to the agent

Instructions loaded
≈57 tok
1 file · 81 in bundle
Tool list per prompt
≈0 tok
0 tools · 0 per tool
Handshake
no reply
launch to tools/list, stdio
Undocumented tools
0
description under 20 chars · avg 0

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 @browserless.io/mcp@1.47.1 as of 8d ago. This is the package itself, not its dependencies.