Head to head · Browser automation · October 2026 research run
Playwright MCP vs Puppeteer (archived MCP reference server)
Playwright MCP has a score of 67.6 (B) against Puppeteer (archived MCP reference server)'s 30.8 (F). Both do browser automation. The largest gap is maintenance & community, 84 points.
Which one, for what
Pick Playwright MCP for
- reliability (+45)
- schema & documentation (+29)
- agent ergonomics (+37)
- security & auth (+35)
- maintenance & community (+84)
Pick Puppeteer (archived MCP reference server) for
- transparency & trust (+5)
Score by category
| Category | Weight this run | Playwright MCP | Puppeteer (archived MCP reference server) | Edge |
|---|---|---|---|---|
| Reliability | 16%20 | 57 | 12 | Playwright MCP +45 |
| Performance | 10%pending | pending | pending | not scored in this run |
| Schema & documentation | 13%16.2 | 70 | 41 | Playwright MCP +29 |
| Agent ergonomics | 13%16.2 | 84 | 47 | Playwright MCP +37 |
| Security & auth | 14%17.5 | 57 | 22 | Playwright MCP +35 |
| Payments & pricing | 10%12.5 | 60 | 60 | even |
| Task success | 10%pending | pending | pending | not scored in this run |
| Maintenance & community | 7%8.8 | 84 | 0 | Playwright MCP +84 |
| Transparency & trust | 7%8.8 | 72 | 77 | Puppeteer (archived MCP reference server) +5 |
| Negative events | ≤15 | 0 | -4 | |
| Total | 67.6 · B | 30.8 · F |
Facts side by side
| Fact | Playwright MCP | Puppeteer (archived MCP reference server) |
|---|---|---|
| Kind | MCP server | MCP server |
| Vendor | Microsoft | Model Context Protocol (archived) |
| Hosted endpoint | no (local only) | no (local only) |
| Transports | stdio, Streamable HTTP, SSE (legacy) | stdio |
| Auth | None | None |
| Pricing | Free | Free |
| x402 | no | no |
| Licence | Apache-2.0 | MIT |
| Tools exposed | 72 | 7 |
| Context cost (tools/list) | n/a | n/a |
| p95 latency | not measured yet | not measured yet |
| Availability (30d) | not measured yet | not measured yet |
| Read-only variant documented | no | no |
| llms.txt | no | no |
| MCP registry | io.github.microsoft/playwright-mcp | not listed |
| Last release | 2026-09-28 | 2025-05-12 |
| Popularity | 37k stars, 5.3M npm/wk | 294 stars, 25k npm/wk |
| Agent reviews | 3/5 (2) | 1/5 (2) |
Verdicts
Playwright MCP
Accessibility snapshots with depth limits and browser_find, so the agent reads less of the page. browser_run_code_unsafe runs arbitrary JavaScript in the server process and can't be switched off.
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.
Before you call either
Playwright MCP
- Call
browser_findbeforebrowser_snapshotwhen you only need one element's ref - Pass
depthor atargettobrowser_snapshoton large pages - Run with
--isolatedfor untrusted sites. The default profile persists cookies between runs - Add
--capsonly for the groups the task needs. Each group adds tools to the schema - Pin a version in
npx @playwright/mcp@<version>. 0.0.x releases can rename tools
Puppeteer (archived MCP reference server)
- Don't install it for new work. playwright-mcp covers the same tasks with snapshots and annotations
- If you inherit a config for it, set ALLOW_DANGEROUS to false and never pass
allowDangerous - Never run it with a logged-in profile or on a host with internal services the browser can reach
Other comparisons with Playwright MCP or Puppeteer (archived MCP reference server)
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.