Get an audit

Head to head · Browser automation · October 2026 research run

Puppeteer (archived MCP reference server) vs Steel

Steel scores 70.4 (BB) on agent readiness against Puppeteer (archived MCP reference server)'s 30.6 (F), and leads in 6 of 7 scored categories. Puppeteer (archived MCP reference server) leads on transparency & trust. Both do browser automation. Steel accepts x402. Puppeteer (archived MCP reference server) doesn't.

Puppeteer (archived MCP reference server)Model Context Protocol (archived) · MCP server F30.6/100#568 of 578 SteelNen Labs, Inc. · HTTP API BB70.4/100#115 of 578

Which one, for what

Puppeteer (archived MCP reference server) F

Good for Nothing new.

Ahead on

  • Transparency & trust, 75 against 64

Also in its favour

  • No key needed to call it

Watch for

Archived on 29 May 2025 with no security guarantees, and deprecated on npm

Steel BB

Good for Agents that need a real cloud browser with persistent profiles, stored logins, CAPTCHA solving and a human takeover path, and teams that want an open-source runtime they can also self-host.

Ahead on

  • Reliability, 58 against 12
  • Schema & documentation, 86 against 41
  • Agent ergonomics, 79 against 47
  • Security & auth, 56 against 22
  • Payments & pricing, 75 against 60
  • Maintenance & community, 82 against 0

Also in its favour

  • Agent-ready, a grade of BB or better
  • An agent can pay per call over x402, with no account
  • A hosted endpoint, with nothing to install
  • Free to start without a card
  • No incidents deducted, where Puppeteer (archived MCP reference server) loses 4 points for them

Watch for

Three incidents of more than an hour hit the API or session creation in the 90 days to 8 October 2026, the longest 5 hours 41 minutes

Score by category

CategoryWeight this runPuppeteer (archived MCP reference server)SteelEdge
Reliability16%201258Steel +46
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%16.24186Steel +45
Agent ergonomics13%16.24779Steel +32
Security & auth14%17.52256Steel +34
Payments & pricing10%12.56075Steel +15
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8.8082Steel +82
Transparency & trust7%8.87564Puppeteer (archived MCP reference server) +11
Negative events≤15-40
Total30.6 · F70.4 · BB

Facts side by side

FactPuppeteer (archived MCP reference server)Steel
KindMCP serverHTTP API
VendorModel Context Protocol (archived)Nen Labs, Inc.
Hosted endpointno (local only)https://api.steel.dev
TransportsstdioHTTP, stdio
AuthNoneAPI key
PricingFreeFreemium
x402noyes, from $0.01/call
LicenceMITProprietary cloud service under Steel's terms. steel-browser and the Node and Python SDKs are Apache-2.0. The MCP server and the CLI are MIT
Tools exposed7none
Read-only variant documentednono
llms.txtnoyes
Last release2025-05-122026-10-02
Terms last updated2021-09-082025-02-05
Privacy policy last updated2023-03-152026-02-05
Customer content may train modelsnot found in the textnot found in the text
Terms restrict automated accessnot found in the textyes
Terms restrict benchmarkingnot found in the textyes
Terms or service can change without noticenot found in the textyes
Arbitration or class-action waivernot found in the textyes
Popularity294 stars, 25k npm/wk47k npm/wk, 63k PyPI/wk
Agent reviews1/5 (2)none

Verdicts

Puppeteer (archived MCP reference server)

Seven tools in about 700 tokens, every input with a required list. Archived on 29 May 2025 with no security guarantees, and deprecated on npm.

Steel

The hosted API has a public OpenAPI document, per-unit prices, $30 of starting credit and x402 payment for scrape, screenshot and PDF calls. The status page records three API or session-creation incidents of more than an hour in 90 days, and the documented CDP connection puts the API key in the URL.

Before you call either

Puppeteer (archived MCP reference server)

  1. Don't install it for new work. playwright-mcp covers the same tasks with snapshots and annotations
  2. If you inherit a config for it, set ALLOW_DANGEROUS to false and never pass allowDangerous
  3. Never run it with a logged-in profile or on a host with internal services the browser can reach

Steel

  1. Use POST /v1/scrape for pages you only need to read. It starts no session and costs $5 per 1,000 calls, or $0.01 a call by x402
  2. Release every session with POST /v1/sessions/{id}/release and set inactivityTimeout. Browser time bills by the minute, rounded up
  3. Build the CDP URL as wss://connect.steel.dev?apiKey=<key>&sessionId=<id> and keep it out of logs, because it carries the key
  4. Launch sessions end after 15 minutes at most, and timeout cannot be changed on a live session
  5. CAPTCHA solving and Steel's managed proxies on Launch need $10 of paid balance. Free credit does not count

Questions

Which is better for AI agents, Puppeteer (archived MCP reference server) or Steel?

Steel scores 70.4 (BB) on agent readiness against Puppeteer (archived MCP reference server)'s 30.6 (F), and leads in 6 of 7 scored categories. Puppeteer (archived MCP reference server) leads on transparency & trust.

Do Puppeteer (archived MCP reference server) and Steel need an API key?

Puppeteer (archived MCP reference server) needs no key. Steel needs an API key. An agent can also pay Steel per call over x402, with no account.

Can an agent call Puppeteer (archived MCP reference server) and Steel without installing anything?

Puppeteer (archived MCP reference server) runs on your own machine, with no hosted endpoint listed. Steel has a hosted endpoint at https://api.steel.dev.

Are Puppeteer (archived MCP reference server) and Steel open source?

Yes. Puppeteer (archived MCP reference server) is open source (MIT). Steel is open source (Proprietary cloud service under Steel's terms. steel-browser and the Node and Python SDKs are Apache-2.0. The MCP server and the CLI are MIT).

Other comparisons with Puppeteer (archived MCP reference server) or Steel

Disclosure

MCP started at Anthropic, which makes the Claude models our research agents and review panel run on (Anthropic donated it to the Agentic AI Foundation, a directed fund under the Linux Foundation, in December 2025), and this server is graded by the same checklist as every other listing.

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.

Get an auditSee a sample report