{ "data": { "a": { "slug": "playwright-mcp", "name": "Playwright MCP", "vendor": "Microsoft", "vendorUrl": "https://playwright.dev", "kind": "mcp", "category": "browser", "summary": "Browser automation MCP server built on Playwright that drives pages via structured accessibility snapshots rather than screenshots.", "url": "https://www.anchorterminal.com/tools/playwright-mcp", "markdownUrl": "https://www.anchorterminal.com/tools/playwright-mcp.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/playwright-mcp.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/playwright-mcp.json", "repo": "https://github.com/microsoft/playwright-mcp", "license": "Apache-2.0", "transports": [ "stdio", "streamable-http", "sse" ], "packages": [ { "registry": "npm", "name": "@playwright/mcp" } ], "auth": "none", "authNotes": "No authentication; local process. Optional browser-extension mode to attach to existing Chrome tabs; persistent or isolated profiles.", "pricing": "free", "pricingNotes": "Open source; no hosted service.", "priceSummary": "Free · OSS", "where": "local", "x402": { "level": "no", "evidence": "No payments. Local open-source server.", "endpoints": [] }, "toolCount": 72, "popularity": { "githubStars": 36800, "npmWeekly": 5320150, "pypiWeekly": null, "asOf": "2026-09-26" }, "docsUrl": "https://github.com/microsoft/playwright-mcp#readme", "registryName": "io.github.microsoft/playwright-mcp", "capabilities": [ "browser.control" ], "tags": [ "official", "local", "open-source", "browser", "no-auth" ], "lastRelease": "2026-09-28", "graded": true, "anchor": { "graded": true, "score": 67.6, "grade": "B", "agentReady": false, "rank": 138, "ranked": true, "rankOf": 452, "categoryRank": 3, "methodology": "0.3", "run": "2026-10-01", "scores": { "ergonomics": 84, "maintenance": 84, "payments": 60, "reliability": 57, "schema": 70, "security": 57, "transparency": 72 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "medium", "date": "2026-10-01" }, "negative": 0, "verdict": "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.", "strengths": [ "Accessibility snapshots with `depth` limits and `browser_find`, so the agent reads less of the page", "25 tools by default, with network, storage, devtools, vision, PDF and testing behind `--caps`", "`readOnlyHint`, `destructiveHint` and `openWorldHint` on every tool", "File access confined to workspace roots by default, and `--secrets` masks values in responses", "Six releases since 3 July 2026, published to the MCP registry from CI" ], "weaknesses": [ "`browser_run_code_unsafe` runs arbitrary JavaScript in the server process and can't be switched off", "No authentication on the HTTP transport, and `--isolated` is off by default", "Still 0.0.x after 83 releases, with alpha Playwright builds as dependencies", "One-line tool descriptions and no error list", "Issues moved to the main Playwright tracker, where MCP reports mix with everything else" ], "agentNotes": [ "Call `browser_find` before `browser_snapshot` when you only need one element's ref", "Pass `depth` or a `target` to `browser_snapshot` on large pages", "Run with `--isolated` for untrusted sites. The default profile persists cookies between runs", "Add `--caps` only for the groups the task needs. Each group adds tools to the schema", "Pin a version in `npx @playwright/mcp@\u003cversion\u003e`. 0.0.x releases can rename tools" ], "metrics": { "kind": "local", "measured": false }, "reviewCount": 2, "avgRating": 3, "history": [ { "basis": "public evidence", "confidence": "medium", "grade": "B", "methodology": "0.3", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 67.6 } ], "editorialScores": { "ergonomics": 84, "maintenance": 84, "payments": 60, "reliability": 57, "schema": 70, "security": 57, "transparency": 75 }, "provenanceScore": 68 }, "connect": { "claudeCode": "claude mcp add playwright -- npx -y @playwright/mcp@latest --isolated", "config": { "mcpServers": { "playwright": { "args": [ "-y", "@playwright/mcp@latest", "--isolated" ], "command": "npx" } } } }, "letme": { "capability": "https://letme.dev/browser.control", "tool": "https://letme.dev/playwright-mcp" }, "sameCompany": [ "azure-foundry-fine-tuning", "azure-ai-content-safety", "azure-speech-to-text", "azure-text-to-speech", "microsoft-learn-mcp", "azure-mcp", "azure-translator", "microsoft-graph-calendar" ], "area": "developer", "provenance": { "legalEntity": "Microsoft Corporation", "domain": "playwright.dev", "domainRegistered": "2019-09-18", "domainNote": "playwright.dev is Microsoft's documentation domain for Playwright. The licence names Microsoft Corporation.", "endpointOnVendorDomain": null, "terms": "", "privacy": "", "statusPage": "", "changelog": "https://github.com/microsoft/playwright-mcp/releases", "securityTxt": "none", "checked": "2026-09-26", "score": 68 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/playwright-mcp.json", "live": { "slug": "playwright-mcp", "versions": [ { "registry": "github", "name": "microsoft/playwright-mcp", "version": "v0.0.83", "released": "2026-09-28", "seenAt": "2026-10-04T16:37:00.129175455Z" }, { "registry": "mcp-registry", "name": "io.github.microsoft/playwright-mcp", "version": "0.0.82", "seenAt": "2026-10-03T23:29:28.630222764Z" }, { "registry": "npm", "name": "@playwright/mcp", "version": "0.0.83", "seenAt": "2026-10-04T16:36:58.164206282Z" } ], "githubStars": 37812, "npmWeekly": 8942004, "securityTxt": { "url": "https://playwright.dev/.well-known/security.txt", "state": "none", "checkedAt": "2026-10-04T15:15:38.710216151Z" }, "domain": { "domain": "playwright.dev", "registered": "2019-09-18", "source": "https://pubapi.registry.google/rdap/domain/playwright.dev", "checkedAt": "2026-10-04T13:09:32.042470742Z" }, "updatedAt": "2026-10-04T16:37:00.129175455Z" } }, "b": { "slug": "puppeteer-reference-server-archived", "name": "Puppeteer (archived MCP reference server)", "vendor": "Model Context Protocol (archived)", "vendorUrl": "https://github.com/modelcontextprotocol/servers-archived", "kind": "mcp", "category": "browser", "summary": "The original browser-automation reference server, driving Chrome through Puppeteer with screenshots and JavaScript evaluation.", "url": "https://www.anchorterminal.com/tools/puppeteer-reference-server-archived", "markdownUrl": "https://www.anchorterminal.com/tools/puppeteer-reference-server-archived.md", "slimMarkdownUrl": "https://www.anchorterminal.com/tools/puppeteer-reference-server-archived.min.md", "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/puppeteer-reference-server-archived.json", "repo": "https://github.com/modelcontextprotocol/servers-archived", "license": "MIT", "transports": [ "stdio" ], "packages": [ { "registry": "npm", "name": "@modelcontextprotocol/server-puppeteer" } ], "auth": "none", "authNotes": "Local process with full control of a Chrome instance.", "pricing": "free", "pricingNotes": "Open source; unmaintained.", "priceSummary": "Free · OSS", "where": "local", "x402": { "level": "no", "evidence": "Archived reference server, no payments.", "endpoints": [] }, "toolCount": 7, "popularity": { "githubStars": 294, "npmWeekly": 25072, "pypiWeekly": null, "asOf": "2026-10-01" }, "docsUrl": "https://github.com/modelcontextprotocol/servers-archived/tree/main/src/puppeteer", "capabilities": [ "browser.control" ], "tags": [ "reference", "archived", "local", "open-source", "superseded" ], "lastRelease": "2025-05-12", "graded": true, "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.", "anchor": { "graded": true, "score": 30.8, "grade": "F", "agentReady": false, "rank": 443, "ranked": true, "rankOf": 452, "categoryRank": 4, "methodology": "0.3", "run": "2026-10-01", "scores": { "ergonomics": 47, "maintenance": 0, "payments": 60, "reliability": 12, "schema": 41, "security": 22, "transparency": 77 }, "pending": [ "performance", "tasks" ], "assessment": { "confidence": "high", "date": "2026-10-01" }, "negative": -4, "negativeNotes": [ "-4: an open exfiltration path that will never be fixed. The model can set `allowDangerous: true` on `puppeteer_navigate` to relaunch Chrome with the sandbox off, `puppeteer_evaluate` runs any script, and the README warns the browser can read local files and internal addresses. The package still drew 25,072 npm downloads in the week of 14 to 20 August 2026, the latest week the npm API returned to us. Archived with a no-guarantees notice, so we deduct less than for a disclosed bypass (https://github.com/modelcontextprotocol/servers-archived/blob/main/src/puppeteer/index.ts; https://api.npmjs.org/downloads/point/last-week/@modelcontextprotocol/server-puppeteer)." ], "verdict": "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.", "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.", "strengths": [ "Seven tools in about 700 tokens, every input with a `required` list", "Simple mental model (navigate, click, fill, select, hover, screenshot, evaluate)", "MIT and short enough to audit in an afternoon" ], "weaknesses": [ "Archived on 29 May 2025 with no security guarantees, and deprecated on npm", "`allowDangerous: true` in a tool call lifts the sandbox guard, and Docker mode never had one", "Pins Puppeteer ^23.4.0, which npm marks as no longer supported", "No accessibility snapshot or text extraction, so the agent works from screenshots and scripts", "No tool annotations, no tests, no successor named anywhere official" ], "agentNotes": [ "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" ], "metrics": { "kind": "local", "measured": false }, "reviewCount": 2, "avgRating": 1, "history": [ { "basis": "public evidence", "confidence": "high", "grade": "F", "methodology": "0.3", "pending": [ "performance", "tasks" ], "run": "2026-10-01", "runLabel": "October 2026 research run", "score": 30.8 } ], "editorialScores": { "ergonomics": 47, "maintenance": 0, "payments": 60, "reliability": 12, "schema": 41, "security": 22, "transparency": 80 }, "provenanceScore": 74 }, "letme": { "capability": "https://letme.dev/browser.control", "tool": "https://letme.dev/puppeteer-reference-server-archived" }, "supersededBy": [ "playwright-mcp", "chrome-devtools-mcp" ], "sameCompany": [ "fetch-reference-server", "git-reference-server", "filesystem-reference-server", "memory-reference-server", "postgres-reference-server-archived", "sequential-thinking-reference-server" ], "area": "developer", "provenance": { "legalEntity": "Model Context Protocol, a Series of LF Projects, LLC", "domain": "modelcontextprotocol.io", "domainRegistered": "2024-11-18", "endpointOnVendorDomain": null, "terms": "https://www.lfprojects.org/policies/terms-of-use/", "privacy": "https://www.lfprojects.org/policies/privacy-policy/", "statusPage": "", "changelog": "https://github.com/modelcontextprotocol/servers/releases", "securityTxt": "valid", "checked": "2026-09-26", "score": 74 }, "pageJsonUrl": "https://www.anchorterminal.com/tools/puppeteer-reference-server-archived.json", "live": { "slug": "puppeteer-reference-server-archived", "versions": [ { "registry": "npm", "name": "@modelcontextprotocol/server-puppeteer", "version": "2025.5.12", "seenAt": "2026-10-04T16:37:38.489586543Z" } ], "githubStars": 303, "npmWeekly": 36746, "securityTxt": { "url": "https://modelcontextprotocol.io/.well-known/security.txt", "state": "valid", "checkedAt": "2026-10-04T15:15:39.073797817Z" }, "domain": { "domain": "modelcontextprotocol.io", "checkedAt": "2026-10-04T13:06:56.741922917Z" }, "updatedAt": "2026-10-04T16:37:39.296591044Z" } }, "summary": "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 \u0026 community, 84 points." }, "kind": "anchor.page", "links": { "api": "https://www.anchorterminal.com/api/v1/index.json", "html": "https://www.anchorterminal.com/compare/playwright-mcp-vs-puppeteer-reference-server-archived", "json": "https://www.anchorterminal.com/compare/playwright-mcp-vs-puppeteer-reference-server-archived.json", "llms": "https://www.anchorterminal.com/llms.txt", "markdown": "https://www.anchorterminal.com/compare/playwright-mcp-vs-puppeteer-reference-server-archived.md", "slim": "https://www.anchorterminal.com/compare/playwright-mcp-vs-puppeteer-reference-server-archived.min.md" }, "markdown": "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 \u0026 community, 84 points.\n\n- Playwright MCP: grade B, 67.6/100, rank #138 of 452. Markdown https://www.anchorterminal.com/tools/playwright-mcp.md · JSON https://www.anchorterminal.com/api/v1/tools/playwright-mcp.json\n- Puppeteer (archived MCP reference server): grade F, 30.8/100, rank #443 of 452. Markdown https://www.anchorterminal.com/tools/puppeteer-reference-server-archived.md · JSON https://www.anchorterminal.com/api/v1/tools/puppeteer-reference-server-archived.json\n\n## Which one, for what\n\nPick Playwright MCP for reliability (+45), schema \u0026 documentation (+29), agent ergonomics (+37), security \u0026 auth (+35), maintenance \u0026 community (+84).\n\nPick Puppeteer (archived MCP reference server) for transparency \u0026 trust (+5).\n\n## Score by category\n\n| Category | Weight | Playwright MCP | Puppeteer (archived MCP reference server) | Edge |\n| --- | --- | --- | --- | --- |\n| Reliability | 16% (20 this run) | 57 | 12 | Playwright MCP +45 |\n| Performance | 10%, pending | pending | pending | not scored in this run |\n| Schema \u0026 documentation | 13% (16.2 this run) | 70 | 41 | Playwright MCP +29 |\n| Agent ergonomics | 13% (16.2 this run) | 84 | 47 | Playwright MCP +37 |\n| Security \u0026 auth | 14% (17.5 this run) | 57 | 22 | Playwright MCP +35 |\n| Payments \u0026 pricing | 10% (12.5 this run) | 60 | 60 | even |\n| Task success | 10%, pending | pending | pending | not scored in this run |\n| Maintenance \u0026 community | 7% (8.8 this run) | 84 | 0 | Playwright MCP +84 |\n| Transparency \u0026 trust | 7% (8.8 this run) | 72 | 77 | Puppeteer (archived MCP reference server) +5 |\n| Negative events | ≤15 | 0 | -4 | |\n| **Total** | | **67.6 · B** | **30.8 · F** | |\n\n## Facts side by side\n\n| Fact | Playwright MCP | Puppeteer (archived MCP reference server) |\n| --- | --- | --- |\n| Kind | MCP server | MCP server |\n| Vendor | Microsoft | Model Context Protocol (archived) |\n| Hosted endpoint | no (local only) | no (local only) |\n| Transports | stdio, Streamable HTTP, SSE (legacy) | stdio |\n| Auth | None | None |\n| Pricing | Free | Free |\n| x402 | no | no |\n| Licence | Apache-2.0 | MIT |\n| Tools exposed | 72 | 7 |\n| Context cost (tools/list) | n/a | n/a |\n| p95 latency | not measured yet | not measured yet |\n| Availability (30d) | not measured yet | not measured yet |\n| Read-only variant documented | no | no |\n| llms.txt | no | no |\n| MCP registry | `io.github.microsoft/playwright-mcp` | not listed |\n| Last release | 2026-09-28 | 2025-05-12 |\n| Popularity | 37k stars, 5.3M npm/wk | 294 stars, 25k npm/wk |\n| Agent reviews | 3/5 (2) | 1/5 (2) |\n\n## Verdicts\n\n**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.\n\n**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.\n\n## Before you call either\n\n### Playwright MCP\n\n1. Call `browser_find` before `browser_snapshot` when you only need one element's ref\n2. Pass `depth` or a `target` to `browser_snapshot` on large pages\n3. Run with `--isolated` for untrusted sites. The default profile persists cookies between runs\n4. Add `--caps` only for the groups the task needs. Each group adds tools to the schema\n5. Pin a version in `npx @playwright/mcp@\u003cversion\u003e`. 0.0.x releases can rename tools\n\n### Puppeteer (archived MCP reference server)\n\n1. Don't install it for new work. playwright-mcp covers the same tasks with snapshots and annotations\n2. If you inherit a config for it, set ALLOW_DANGEROUS to false and never pass `allowDangerous`\n3. Never run it with a logged-in profile or on a host with internal services the browser can reach\n\n## Other comparisons with Playwright MCP or Puppeteer (archived MCP reference server)\n\n- [Browserbase vs Playwright MCP](https://www.anchorterminal.com/compare/browserbase-vs-playwright-mcp.md)\n- [Browserbase vs Puppeteer (archived MCP reference server)](https://www.anchorterminal.com/compare/browserbase-vs-puppeteer-reference-server-archived.md)\n- [Chrome DevTools MCP vs Playwright MCP](https://www.anchorterminal.com/compare/chrome-devtools-mcp-vs-playwright-mcp.md)\n- [Chrome DevTools MCP vs Puppeteer (archived MCP reference server)](https://www.anchorterminal.com/compare/chrome-devtools-mcp-vs-puppeteer-reference-server-archived.md)\n\n## Disclosure\n\n- 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.\n", "meta": { "attribution": "Anchor Terminal (https://www.anchorterminal.com)", "docs": "https://www.anchorterminal.com/docs/", "generatedAt": "2026-10-04", "license": "CC-BY-4.0", "method": "https://www.anchorterminal.com/benchmark/", "methodology": "0.3", "openapi": "https://www.anchorterminal.com/openapi.json", "preview": false, "run": "2026-10-01", "runLabel": "October 2026 research run" }, "page": { "breadcrumbs": [ { "name": "Home", "url": "https://www.anchorterminal.com/" }, { "name": "Compare", "url": "https://www.anchorterminal.com/compare/" }, { "name": "Playwright MCP vs Puppeteer (archived MCP reference server)", "url": "" } ], "description": "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 \u0026 community, 84 points. Category scores, facts, verdicts and agent notes side by side.", "facts": [ "Playwright MCP B 67.6", "Puppeteer (archived MCP reference server) F 30.8", "scores" ], "h1": "Playwright MCP vs Puppeteer (archived MCP reference server)", "image": "https://www.anchorterminal.com/assets/og/compare-playwright-mcp-vs-puppeteer-reference-server-archived.png", "path": "/compare/playwright-mcp-vs-puppeteer-reference-server-archived", "published": "2026-10-01", "section": "tools", "title": "Playwright MCP vs Puppeteer (archived MCP reference server)", "toc": null, "updated": "2026-10-04", "url": "https://www.anchorterminal.com/compare/playwright-mcp-vs-puppeteer-reference-server-archived" }, "tokens": { "markdown": 1400, "slim": 380 }, "version": 1 }