{ "data": { "category": { "area": "developer", "capabilities": [ "browser.control", "browser.debug" ], "description": "Servers that let an agent drive a real browser. Navigate, read accessibility trees, click, fill forms, capture traces and screenshots.", "indexed": [ { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/agentutility-mcp-browser-workflow.json", "kind": "mcp", "name": "AgentUtility Browser Workflow", "slug": "agentutility-mcp-browser-workflow", "url": "https://www.anchorterminal.com/tools/agentutility-mcp-browser-workflow" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/aginxbrowser.json", "kind": "mcp", "name": "Aginx Browser", "slug": "aginxbrowser", "url": "https://www.anchorterminal.com/tools/aginxbrowser" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/appcontext.json", "kind": "mcp", "name": "AppContext", "slug": "appcontext", "url": "https://www.anchorterminal.com/tools/appcontext" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/e2llm-element-to-llm.json", "kind": "mcp", "name": "E2LLM", "slug": "e2llm-element-to-llm", "url": "https://www.anchorterminal.com/tools/e2llm-element-to-llm" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/foura-mcp.json", "kind": "mcp", "name": "foura.ai MCP server", "slug": "foura-mcp", "url": "https://www.anchorterminal.com/tools/foura-mcp" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/gamedev-creator.json", "kind": "mcp", "name": "gamedev.pl", "slug": "gamedev-creator", "url": "https://www.anchorterminal.com/tools/gamedev-creator" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/sylin-ghostlight.json", "kind": "mcp", "name": "ghostlight", "slug": "sylin-ghostlight", "url": "https://www.anchorterminal.com/tools/sylin-ghostlight" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/openbrowser-ai.json", "kind": "mcp", "name": "openbrowser-ai", "slug": "openbrowser-ai", "url": "https://www.anchorterminal.com/tools/openbrowser-ai" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/pilotgentic-mcp-server.json", "kind": "mcp", "name": "PilotGentic", "slug": "pilotgentic-mcp-server", "url": "https://www.anchorterminal.com/tools/pilotgentic-mcp-server" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/playrunner-mcp.json", "kind": "mcp", "name": "Playrunner", "slug": "playrunner-mcp", "url": "https://www.anchorterminal.com/tools/playrunner-mcp" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/presio.json", "kind": "mcp", "name": "Presio", "slug": "presio", "url": "https://www.anchorterminal.com/tools/presio" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/quickpeek.json", "kind": "mcp", "name": "QuickPeek", "slug": "quickpeek", "url": "https://www.anchorterminal.com/tools/quickpeek" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/reflexmcp-reflex.json", "kind": "mcp", "name": "reflex", "slug": "reflexmcp-reflex", "url": "https://www.anchorterminal.com/tools/reflexmcp-reflex" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/screenbrowser.json", "kind": "mcp", "name": "Screen Browser", "slug": "screenbrowser", "url": "https://www.anchorterminal.com/tools/screenbrowser" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/shardflux-mcp.json", "kind": "mcp", "name": "Shardflux", "slug": "shardflux-mcp", "url": "https://www.anchorterminal.com/tools/shardflux-mcp" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/shiplight.json", "kind": "mcp", "name": "Shiplight", "slug": "shiplight", "url": "https://www.anchorterminal.com/tools/shiplight" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/taada-tada.json", "kind": "mcp", "name": "Tada: automated QA for AI-written code", "slug": "taada-tada", "url": "https://www.anchorterminal.com/tools/taada-tada" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/turnsignal.json", "kind": "mcp", "name": "TurnSignal", "slug": "turnsignal", "url": "https://www.anchorterminal.com/tools/turnsignal" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/nyx-intelligence-val.json", "kind": "mcp", "name": "val", "slug": "nyx-intelligence-val", "url": "https://www.anchorterminal.com/tools/nyx-intelligence-val" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/walkeros-source-browser.json", "kind": "mcp", "name": "walkerOS Source Browser", "slug": "walkeros-source-browser", "url": "https://www.anchorterminal.com/tools/walkeros-source-browser" }, { "jsonUrl": "https://www.anchorterminal.com/api/v1/tools/atagon-kogiqa-mcp.json", "kind": "mcp", "name": "Web Browser (kogiQA)", "slug": "atagon-kogiqa-mcp", "url": "https://www.anchorterminal.com/tools/atagon-kogiqa-mcp" } ], "indexedCount": 21, "json": "https://www.anchorterminal.com/categories/browser.json", "name": "Browser automation", "slug": "browser", "test": "", "title": "Browser automation tools for AI agents", "toolCount": 14, "tools": [ "browser-use", "chrome-devtools-mcp", "browserbase", "browserless", "steel", "tinyfish", "playwright-mcp", "anchor-browser", "hyperbrowser", "agent-browser", "notte", "skyvern", "airtop", "puppeteer-reference-server-archived" ], "url": "https://www.anchorterminal.com/categories/browser" }, "faq": [ { "answer": "Browser Use has the highest benchmark score of the 14 ranked browser automation tools, 77.9 (BB). Chrome DevTools MCP is second with 77 (BB).", "question": "What are the highest-rated browser automation tools for AI agents?" }, { "answer": "5 of the 14 ranked here grade BB or better, the bar for agent-ready on the Anchor benchmark.", "question": "How many browser automation tools are agent-ready?" }, { "answer": "Browser Use, Browserbase, Steel and Hyperbrowser. An agent can pay these per call in USDC with no account.", "question": "Which browser automation tools accept x402 payments?" }, { "answer": "By published paid prices, Browser Use, at $0.02 per browser hour, the lowest of the 7 listings here with a paid price in this unit (free allowances aside). Plans, volume tiers and free allowances change the sum, so check the listing's price table.", "question": "Which of these browser automation tools is cheapest?" }, { "answer": "By the Anchor benchmark score out of 100, a weighted mean of the scored categories minus deductions for negative events, from public evidence re-checked as vendors change. Listings cannot pay for a place. The latest assessment behind this page is from 8 October 2026.", "question": "How is this list ranked?" } ], "howToChoose": [ { "label": "Browser lifetime and session reuse", "detail": "Check how long a browser session lives and whether it can be resumed, because an agent that loses its page state mid-task must restart the whole flow." }, { "label": "Credentials for signed-in sites", "detail": "Check where site logins are stored and who can read them, since an agent acting in a signed-in browser can reach anything that session can." }, { "label": "Accessibility tree and page reads", "detail": "Check whether the tool returns an accessibility tree or only raw HTML and screenshots, since a structured tree lets the agent find and fill fields without parsing markup." }, { "label": "Traces and screenshots on failure", "detail": "Check whether traces and screenshots are saved for failed runs and how long they are kept, since an agent that cannot see why a step failed will retry blindly." } ], "picks": [ { "also": { "name": "Chrome DevTools MCP", "slug": "chrome-devtools-mcp", "why": "BB, 77/100" }, "name": "Browser Use", "need": "Highest score overall", "slug": "browser-use", "why": "BB, 77.9/100 on the benchmark" }, { "name": "Browserbase", "need": "Reliability", "slug": "browserbase", "why": "90/100 on reliability, against 87 for the overall leader" }, { "name": "Browserless", "need": "Schema \u0026 documentation", "slug": "browserless", "why": "89/100 on schema \u0026 documentation, against 85 for the overall leader" }, { "name": "Playwright MCP", "need": "Agent ergonomics", "slug": "playwright-mcp", "why": "84/100 on agent ergonomics, against 78 for the overall leader" }, { "name": "agent-browser", "need": "Security \u0026 auth", "slug": "agent-browser", "why": "67/100 on security \u0026 auth, against 59 for the overall leader" }, { "name": "Chrome DevTools MCP", "need": "Maintenance \u0026 community", "slug": "chrome-devtools-mcp", "why": "93/100 on maintenance \u0026 community, against 77 for the overall leader" }, { "name": "Chrome DevTools MCP", "need": "Transparency \u0026 trust", "slug": "chrome-devtools-mcp", "why": "81/100 on transparency \u0026 trust, against 65 for the overall leader" }, { "also": { "name": "Anchor Browser", "slug": "anchor-browser", "why": "$0.05 per browser hour" }, "name": "Browser Use", "need": "Lowest paid price per browser hour", "slug": "browser-use", "why": "$0.02 per browser hour, the lowest of the 7 listings here with a paid price in this unit (free allowances aside)" }, { "also": { "name": "Browserbase", "slug": "browserbase", "why": "accepts x402" }, "name": "Browser Use", "need": "Paying per call with no account (x402)", "slug": "browser-use", "why": "accepts x402" }, { "also": { "name": "Steel", "slug": "steel", "why": "self-hosted, Proprietary cloud service under Steel's terms licence" }, "name": "Browserless", "need": "Self-hosting under an open licence", "slug": "browserless", "why": "self-hosted, SSPL 1 licence" }, { "name": "Chrome DevTools MCP", "need": "The review panel's favourite", "slug": "chrome-devtools-mcp", "why": "3.5/5 from 8 panel reviews" } ], "ranked": 14, "shortlist": [ { "bestFor": "An agent that wants to hand a whole browser task to a hosted agent and get a result back, or that needs a low-cost cloud browser over CDP, and for unattended agents that can pay by x402.", "grade": "BB", "name": "Browser Use", "position": 1, "price": "$0.02 / browser-hr", "score": 77.9, "slug": "browser-use", "strengths": [ "x402 host at x402.api.browser-use.com takes USDC on Base with no account or API key, and wallet top-ups start at $0.01", "Public prices without login, $0.02 a browser-hour, proxy traffic $5 a GB, model tokens plus a 20 per cent fee", "OpenAPI 3.1.0 documents for API v2, v3 and v4, llms.txt and Markdown copies of every docs page" ], "url": "https://www.anchorterminal.com/tools/browser-use", "verdict": "The hosted cloud accepts x402 top-ups in USDC on Base with no account or API key, and publishes prices, rate limits and three OpenAPI documents. The privacy policy says inputs are used to train Browser Use's models, API keys have no documented scopes, and no prompt-injection guidance was found in the docs.", "weaknesses": [ "The privacy policy says inputs are used to train Browser Use's AI models, with no opt-out described outside Enterprise zero data retention", "No scopes or permissions documented for API keys, and the per-key monthly spending cap is a soft limit", "No prompt-injection guidance found in the cloud docs, though agents read untrusted pages" ], "where": "both", "x402": "yes" }, { "bestFor": "Coding agents debugging or profiling a web app they're building.", "grade": "BB", "name": "Chrome DevTools MCP", "position": 2, "price": "Free · OSS", "score": 77, "slug": "chrome-devtools-mcp", "strengths": [ "Performance traces, network inspection, heap snapshots and Lighthouse audits in one server", "Every tool carries `readOnlyHint`, and category flags and `--slim` cut the tool list from about 30 to three", "`--javascript-evaluation false`, URL allow and block patterns and MCP roots for least privilege" ], "url": "https://www.anchorterminal.com/tools/chrome-devtools-mcp", "verdict": "Performance traces, network inspection, heap snapshots and Lighthouse audits in one server. Usage statistics go to Google by default until you pass `--no-usage-statistics`.", "weaknesses": [ "Usage statistics go to Google by default until you pass `--no-usage-statistics`", "`--isolated` and network header redaction are off by default, so the agent sees a persistent profile and raw headers", "1.8.0 made `pageId` required on page tools in a minor release, filed under `Features`" ], "where": "local", "x402": "no" }, { "bestFor": "Agents that need a real JavaScript-capable browser in the cloud, especially unattended ones that can pay by x402.", "grade": "BB", "name": "Browserbase", "position": 3, "price": "$0.12 / browser-hr", "score": 76.2, "slug": "browserbase", "strengths": [ "x402 sessions at x402.browserbase.com with no account, $0.12 an hour, unused minutes refunded", "Per-plan concurrency and session-creation limits, with Retry-After and a documented backoff helper", "OpenAPI 3.0.0, llms.txt and a dated changelog with 12 entries since 13 July 2026" ], "url": "https://www.anchorterminal.com/tools/browserbase", "verdict": "x402 sessions at x402.browserbase.com with no account, $0.12 an hour, unused minutes refunded. The hosted MCP setup page passes the API key as `?browserbaseApiKey=` in the URL.", "weaknesses": [ "The hosted MCP setup page passes the API key as `?browserbaseApiKey=` in the URL", "No scopes or rotation guide for API keys, and no SLA published", "Hosted MCP tools have one-line descriptions and no annotations" ], "where": "hosted", "x402": "yes" }, { "bestFor": "Agents that need a real browser in the cloud with stealth, proxies and CAPTCHA solving, and teams that want the option of running the same server themselves.", "grade": "BB", "name": "Browserless", "position": 4, "price": "$0.002 / credit", "score": 70.4, "slug": "browserless", "strengths": [ "OpenAPI 3.0.0 document with 87 paths, llms.txt, llms-full.txt and Markdown twins of the docs", "Hosted MCP server at mcp.browserless.io with 14 documented tools, each carrying read-only and destructive annotations", "Free plan with 1,000 units a month and no card, with per-unit overage prices published for paid plans" ], "url": "https://www.anchorterminal.com/tools/browserless", "verdict": "An OpenAPI document, llms.txt and a 14-tool MCP server with annotations on every tool make the service easy for an agent to read, and the free plan needs no card. The API token travels in the URL query string by default, cloud tokens carry no scopes, and no machine payment route exists.", "weaknesses": [ "REST and WebSocket routes take the API token as `?token=` in the URL, which the docs give as the standard method", "Cloud tokens have no scopes or roles. Token roles exist only on self-hosted Enterprise containers", "No x402, MPP or L402, and an account needs a person to sign up in a browser" ], "where": "both", "x402": "no" }, { "bestFor": "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.", "grade": "BB", "name": "Steel", "position": 5, "price": "$0.01 / call", "score": 70.4, "slug": "steel", "strengths": [ "OpenAPI 3.0.3 document with 75 operations, llms.txt and a Markdown twin of every docs page", "x402 at x402.steel.dev for scrape, screenshot and PDF at $0.01 a call in USDC on Base or Solana, with no account", "Per-unit prices published without login, and $30 of one-time credit on the Launch plan" ], "url": "https://www.anchorterminal.com/tools/steel", "verdict": "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.", "weaknesses": [ "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", "The documented CDP connection passes the API key as `apiKey` in the `wss://connect.steel.dev` URL", "x402 covers the three one-shot endpoints only, so a browser session needs an account and a key" ], "where": "both", "x402": "yes" }, { "bestFor": "Agents that want one account for a goal-driven web agent, a raw CDP browser, and free search and fetch.", "grade": "B", "name": "TinyFish", "position": 6, "price": "$0.016 / call", "score": 67.7, "slug": "tinyfish", "strengths": [ "OpenAPI 3.0 documents for the Agent, Browser, Fetch, Search and Research APIs, plus `llms.txt`, `llms-full.txt` and a Markdown twin of every docs page", "Search and Fetch are free up to 12,000 requests and 1,000 URLs a day, and new accounts start with $8 of wallet funds with no card", "Errors carry a code, a message, optional `details` and a `request_id`. Search and Fetch 429s carry `Retry-After` and `X-RateLimit-*` headers" ], "url": "https://www.anchorterminal.com/tools/tinyfish", "verdict": "OpenAPI documents are published for five APIs, errors carry typed codes and request IDs, and Search and Fetch are free within daily allowances. The terms let Tiny Fish train its models on customer data, run creation takes no idempotency keys, and the status page shows an hour or more of downtime on several days for the Browser and Search APIs.", "weaknesses": [ "The terms of 17 August 2026 allow Tiny Fish to train and improve its models on customer data. No opt-out was found in the reviewed pages", "The status page shows downtime of an hour or more on six days for the Browser API and five for the Search API between 11 July and 8 October 2026, with no incident write-ups", "No idempotency keys on run or session creation. The docs say retrying a failed request may create duplicate runs" ], "where": "hosted", "x402": "no" }, { "bestFor": "The default choice for an agent that drives web pages by structure rather than pixels. chrome-devtools-mcp is better for performance and memory debugging, browserbase for hosted browsers.", "grade": "B", "name": "Playwright MCP", "position": 7, "price": "Free · OSS", "score": 67.6, "slug": "playwright-mcp", "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" ], "url": "https://www.anchorterminal.com/tools/playwright-mcp", "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.", "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" ], "where": "local", "x402": "no" }, { "bestFor": "Agents that must work inside signed-in accounts, with managed identities, MFA handling and a human sign-in link, and agents that need to start with no human signup.", "grade": "B", "name": "Anchor Browser", "position": 8, "price": "$0.01 / call", "score": 67.3, "slug": "anchor-browser", "strengths": [ "Agent Access issues an API key through two unauthenticated calls and a puzzle, with 1 trial credit, or 5 with a verified-email OIDC token", "OpenAPI 3.1 document with 101 paths and 136 operations, llms.txt, llms-full.txt and a Markdown twin of every docs page", "Per-unit prices published without login, $0.01 a browser, $0.05 a browser-hour and $8 a GB of proxy traffic" ], "url": "https://www.anchorterminal.com/tools/anchor-browser", "verdict": "The API has a public OpenAPI 3.1 document, per-unit prices and Agent Access, an unauthenticated flow that issues a trial API key without a dashboard account. No service agreement is published, only a website terms of use, and the documented CDP connection puts the API key in the URL.", "weaknesses": [ "No service agreement or API terms found. anchorbrowser.io/terms is a website terms of use dated 16 March 2025", "The documented CDP connection passes the API key as `apiKey` in the `wss://connect.anchorbrowser.io` URL", "No changelog, deprecation policy or published SLA found, and docs.anchorbrowser.io/changelog returns 404" ], "where": "hosted", "x402": "no" }, { "bestFor": "Agents that need a cloud browser by the minute plus cheap fetch and search, especially wallet-holding agents with no account.", "grade": "B", "name": "Hyperbrowser", "position": 9, "price": "$0.001 / call", "score": 66.9, "slug": "hyperbrowser", "strengths": [ "x402 on /x402/session, /x402/web/fetch and /x402/web/search with USDC on Base or Solana and no account, confirmed by a live 402 on 8 October 2026", "Unit prices public in the docs, $0.10 a browser-hour billed per second, $0.001 a Fetch page, $0.005 a Search query", "Free plan with 5,000 credits and no card, per the pricing page" ], "url": "https://www.anchorterminal.com/tools/hyperbrowser", "verdict": "Sessions, Fetch and Search can be bought by x402 at api.hyperbrowser.ai with USDC on Base or Solana and no account, and unit prices are public. The official MCP server was last published in April 2025, no request rate limits or SLA are published, and the privacy notice covers only the website.", "weaknesses": [ "The official MCP server's last npm release is 1.0.25 of 13 April 2025, and it isn't in the official MCP registry", "No request rate limits, Retry-After header or SLA found; the pricing page says paid plans have no rate limits", "One `x-api-key` per account with no scopes or rotation guidance found in the reviewed documentation" ], "where": "both", "x402": "yes" }, { "bestFor": "Coding agents with a shell that want short commands and compact snapshots.", "grade": "B", "name": "agent-browser", "position": 10, "price": "Free · OSS", "score": 66.6, "slug": "agent-browser", "strengths": [ "Accessibility-tree snapshots give each element a ref such as `@e2`, with `-i`, `-c`, `-d` and `--delta` to cut the output", "The stdio MCP server loads 29 tools by default and 158 with `--tools all`, each with a typed schema and `readOnlyHint`", "Domain allowlist, action policy, action confirmation and content boundary markers are documented with a threat model and known limitations" ], "url": "https://www.anchorterminal.com/tools/agent-browser", "verdict": "Snapshots with element refs, a 29-tool default MCP profile and opt-in domain allowlists, action policies and content boundary markers suit agent use. The project is at 0.38.2 with no stable release, the ten most recent CI runs on main failed between 1 and 8 October 2026, and every security control is off by default.", "weaknesses": [ "The ten most recent CI runs on main, 1 to 8 October 2026, all concluded in failure. In the newest, Windows Rust tests and native end-to-end tests failed", "Version 0.38.2 with a Vercel Labs badge and no stable release or deprecation policy", "All security controls are opt-in. By default there is no limit on navigation, actions or output" ], "where": "local", "x402": "no" } ], "updated": "2026-10-08" }, "kind": "anchor.page", "links": { "api": "https://www.anchorterminal.com/api/v1/index.json", "html": "https://www.anchorterminal.com/best/browser/", "json": "https://www.anchorterminal.com/best/browser/index.json", "llms": "https://www.anchorterminal.com/llms.txt", "markdown": "https://www.anchorterminal.com/best/browser/index.md", "slim": "https://www.anchorterminal.com/best/browser/index.min.md" }, "markdown": "The 10 highest-scoring of 14 browser automation tools on the Anchor benchmark, with a pick for each need and where each one falls short. Scores come from public evidence, re-checked as vendors change.\n\n- Ranked: 14 · agent-ready (BB or better): 5 · accept x402: 4 · hosted endpoints: 10\n- Full ranked table: https://www.anchorterminal.com/categories/browser.md\n- Head-to-head comparisons: https://www.anchorterminal.com/compare/browser/index.md (91)\n- Methodology: https://www.anchorterminal.com/benchmark/index.md\n\n## The shortlist\n\n| # | Tool | Grade | Score | Best for | Price | Where |\n| --- | --- | --- | --- | --- | --- | --- |\n| 1 | [Browser Use](https://www.anchorterminal.com/tools/browser-use.md) | BB | 77.9 | An agent that wants to hand a whole browser task to a hosted agent and get a result back, or that needs a low-cost cloud browser over CDP, and for unattended agents that can pay by x402. | $0.02 / browser-hr | hosted and local |\n| 2 | [Chrome DevTools MCP](https://www.anchorterminal.com/tools/chrome-devtools-mcp.md) | BB | 77 | Coding agents debugging or profiling a web app they're building. | Free · OSS | local |\n| 3 | [Browserbase](https://www.anchorterminal.com/tools/browserbase.md) | BB | 76.2 | Agents that need a real JavaScript-capable browser in the cloud, especially unattended ones that can pay by x402. | $0.12 / browser-hr | hosted |\n| 4 | [Browserless](https://www.anchorterminal.com/tools/browserless.md) | BB | 70.4 | Agents that need a real browser in the cloud with stealth, proxies and CAPTCHA solving, and teams that want the option of running the same server themselves. | $0.002 / credit | hosted and local |\n| 5 | [Steel](https://www.anchorterminal.com/tools/steel.md) | BB | 70.4 | 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. | $0.01 / call | hosted and local |\n| 6 | [TinyFish](https://www.anchorterminal.com/tools/tinyfish.md) | B | 67.7 | Agents that want one account for a goal-driven web agent, a raw CDP browser, and free search and fetch. | $0.016 / call | hosted |\n| 7 | [Playwright MCP](https://www.anchorterminal.com/tools/playwright-mcp.md) | B | 67.6 | The default choice for an agent that drives web pages by structure rather than pixels. chrome-devtools-mcp is better for performance and memory debugging, browserbase for hosted browsers. | Free · OSS | local |\n| 8 | [Anchor Browser](https://www.anchorterminal.com/tools/anchor-browser.md) | B | 67.3 | Agents that must work inside signed-in accounts, with managed identities, MFA handling and a human sign-in link, and agents that need to start with no human signup. | $0.01 / call | hosted |\n| 9 | [Hyperbrowser](https://www.anchorterminal.com/tools/hyperbrowser.md) | B | 66.9 | Agents that need a cloud browser by the minute plus cheap fetch and search, especially wallet-holding agents with no account. | $0.001 / call | hosted and local |\n| 10 | [agent-browser](https://www.anchorterminal.com/tools/agent-browser.md) | B | 66.6 | Coding agents with a shell that want short commands and compact snapshots. | Free · OSS | local |\n\n## Picks by need\n\n- Highest score overall: [Browser Use](https://www.anchorterminal.com/tools/browser-use.md), BB, 77.9/100 on the benchmark. Also [Chrome DevTools MCP](https://www.anchorterminal.com/tools/chrome-devtools-mcp.md), BB, 77/100.\n- Reliability: [Browserbase](https://www.anchorterminal.com/tools/browserbase.md), 90/100 on reliability, against 87 for the overall leader.\n- Schema \u0026 documentation: [Browserless](https://www.anchorterminal.com/tools/browserless.md), 89/100 on schema \u0026 documentation, against 85 for the overall leader.\n- Agent ergonomics: [Playwright MCP](https://www.anchorterminal.com/tools/playwright-mcp.md), 84/100 on agent ergonomics, against 78 for the overall leader.\n- Security \u0026 auth: [agent-browser](https://www.anchorterminal.com/tools/agent-browser.md), 67/100 on security \u0026 auth, against 59 for the overall leader.\n- Maintenance \u0026 community: [Chrome DevTools MCP](https://www.anchorterminal.com/tools/chrome-devtools-mcp.md), 93/100 on maintenance \u0026 community, against 77 for the overall leader.\n- Transparency \u0026 trust: [Chrome DevTools MCP](https://www.anchorterminal.com/tools/chrome-devtools-mcp.md), 81/100 on transparency \u0026 trust, against 65 for the overall leader.\n- Lowest paid price per browser hour: [Browser Use](https://www.anchorterminal.com/tools/browser-use.md), $0.02 per browser hour, the lowest of the 7 listings here with a paid price in this unit (free allowances aside). Also [Anchor Browser](https://www.anchorterminal.com/tools/anchor-browser.md), $0.05 per browser hour.\n- Paying per call with no account (x402): [Browser Use](https://www.anchorterminal.com/tools/browser-use.md), accepts x402. Also [Browserbase](https://www.anchorterminal.com/tools/browserbase.md), accepts x402.\n- Self-hosting under an open licence: [Browserless](https://www.anchorterminal.com/tools/browserless.md), self-hosted, SSPL 1 licence. Also [Steel](https://www.anchorterminal.com/tools/steel.md), self-hosted, Proprietary cloud service under Steel's terms licence.\n- The review panel's favourite: [Chrome DevTools MCP](https://www.anchorterminal.com/tools/chrome-devtools-mcp.md), 3.5/5 from 8 panel reviews.\n\n## How to choose\n\n- Browser lifetime and session reuse: Check how long a browser session lives and whether it can be resumed, because an agent that loses its page state mid-task must restart the whole flow.\n- Credentials for signed-in sites: Check where site logins are stored and who can read them, since an agent acting in a signed-in browser can reach anything that session can.\n- Accessibility tree and page reads: Check whether the tool returns an accessibility tree or only raw HTML and screenshots, since a structured tree lets the agent find and fill fields without parsing markup.\n- Traces and screenshots on failure: Check whether traces and screenshots are saved for failed runs and how long they are kept, since an agent that cannot see why a step failed will retry blindly.\n\n## Each one in detail\n\n### 1. Browser Use, BB 77.9/100\n\nBrowser Use runs AI agents that operate a web browser. It has an MIT Python library and CLI, and a hosted cloud with an agent API, managed browsers over CDP and an MCP server.\n\n- Verdict: The hosted cloud accepts x402 top-ups in USDC on Base with no account or API key, and publishes prices, rate limits and three OpenAPI documents. The privacy policy says inputs are used to train Browser Use's models, API keys have no documented scopes, and no prompt-injection guidance was found in the docs.\n- Choose it for: An agent that wants to hand a whole browser task to a hosted agent and get a result back, or that needs a low-cost cloud browser over CDP, and for unattended agents that can pay by x402.\n- Strength: x402 host at x402.api.browser-use.com takes USDC on Base with no account or API key, and wallet top-ups start at $0.01\n- Strength: Public prices without login, $0.02 a browser-hour, proxy traffic $5 a GB, model tokens plus a 20 per cent fee\n- Strength: OpenAPI 3.1.0 documents for API v2, v3 and v4, llms.txt and Markdown copies of every docs page\n- Weakness: The privacy policy says inputs are used to train Browser Use's AI models, with no opt-out described outside Enterprise zero data retention\n- Weakness: No scopes or permissions documented for API keys, and the per-key monthly spending cap is a soft limit\n- Weakness: No prompt-injection guidance found in the cloud docs, though agents read untrusted pages\n- Price: $0.02 / browser-hr · Auth: API key · x402: yes · Where: hosted and local\n- Full assessment: https://www.anchorterminal.com/tools/browser-use.md\n\n### 2. Chrome DevTools MCP, BB 77/100\n\nLets coding agents control and inspect a live Chrome instance.\n\n- Verdict: Performance traces, network inspection, heap snapshots and Lighthouse audits in one server. Usage statistics go to Google by default until you pass `--no-usage-statistics`.\n- Choose it for: Coding agents debugging or profiling a web app they're building.\n- Strength: Performance traces, network inspection, heap snapshots and Lighthouse audits in one server\n- Strength: Every tool carries `readOnlyHint`, and category flags and `--slim` cut the tool list from about 30 to three\n- Strength: `--javascript-evaluation false`, URL allow and block patterns and MCP roots for least privilege\n- Weakness: Usage statistics go to Google by default until you pass `--no-usage-statistics`\n- Weakness: `--isolated` and network header redaction are off by default, so the agent sees a persistent profile and raw headers\n- Weakness: 1.8.0 made `pageId` required on page tools in a minor release, filed under `Features`\n- Price: Free · OSS · Auth: None · x402: no · Where: local\n- Full assessment: https://www.anchorterminal.com/tools/chrome-devtools-mcp.md\n- Against #1: https://www.anchorterminal.com/compare/browser-use-vs-chrome-devtools-mcp.md\n\n### 3. Browserbase, BB 76.2/100\n\nHosted headless browsers for agents over CDP, plus Fetch and Search APIs and the Stagehand framework.\n\n- Verdict: x402 sessions at x402.browserbase.com with no account, $0.12 an hour, unused minutes refunded. The hosted MCP setup page passes the API key as `?browserbaseApiKey=` in the URL.\n- Choose it for: Agents that need a real JavaScript-capable browser in the cloud, especially unattended ones that can pay by x402.\n- Strength: x402 sessions at x402.browserbase.com with no account, $0.12 an hour, unused minutes refunded\n- Strength: Per-plan concurrency and session-creation limits, with Retry-After and a documented backoff helper\n- Strength: OpenAPI 3.0.0, llms.txt and a dated changelog with 12 entries since 13 July 2026\n- Weakness: The hosted MCP setup page passes the API key as `?browserbaseApiKey=` in the URL\n- Weakness: No scopes or rotation guide for API keys, and no SLA published\n- Weakness: Hosted MCP tools have one-line descriptions and no annotations\n- Price: $0.12 / browser-hr · Auth: API key · x402: yes · Where: hosted\n- Full assessment: https://www.anchorterminal.com/tools/browserbase.md\n- Against #1: https://www.anchorterminal.com/compare/browser-use-vs-browserbase.md\n\n### 4. Browserless, BB 70.4/100\n\nBrowserless runs hosted headless browsers that agents drive over CDP, Playwright, REST endpoints, BrowserQL or its official MCP server. The same software can be self-hosted from a Docker image.\n\n- Verdict: An OpenAPI document, llms.txt and a 14-tool MCP server with annotations on every tool make the service easy for an agent to read, and the free plan needs no card. The API token travels in the URL query string by default, cloud tokens carry no scopes, and no machine payment route exists.\n- Choose it for: Agents that need a real browser in the cloud with stealth, proxies and CAPTCHA solving, and teams that want the option of running the same server themselves.\n- Strength: OpenAPI 3.0.0 document with 87 paths, llms.txt, llms-full.txt and Markdown twins of the docs\n- Strength: Hosted MCP server at mcp.browserless.io with 14 documented tools, each carrying read-only and destructive annotations\n- Strength: Free plan with 1,000 units a month and no card, with per-unit overage prices published for paid plans\n- Weakness: REST and WebSocket routes take the API token as `?token=` in the URL, which the docs give as the standard method\n- Weakness: Cloud tokens have no scopes or roles. Token roles exist only on self-hosted Enterprise containers\n- Weakness: No x402, MPP or L402, and an account needs a person to sign up in a browser\n- Price: $0.002 / credit · Auth: OAuth or key · x402: no · Where: hosted and local\n- Full assessment: https://www.anchorterminal.com/tools/browserless.md\n- Against #1: https://www.anchorterminal.com/compare/browser-use-vs-browserless.md\n\n### 5. Steel, BB 70.4/100\n\nSteel runs cloud browser sessions for AI agents, controlled through a REST API, a CDP WebSocket, SDKs, a CLI and an official MCP server. Its browser runtime, steel-browser, is open source and can be self-hosted.\n\n- Verdict: 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.\n- Choose it 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.\n- Strength: OpenAPI 3.0.3 document with 75 operations, llms.txt and a Markdown twin of every docs page\n- Strength: x402 at x402.steel.dev for scrape, screenshot and PDF at $0.01 a call in USDC on Base or Solana, with no account\n- Strength: Per-unit prices published without login, and $30 of one-time credit on the Launch plan\n- Weakness: 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\n- Weakness: The documented CDP connection passes the API key as `apiKey` in the `wss://connect.steel.dev` URL\n- Weakness: x402 covers the three one-shot endpoints only, so a browser session needs an account and a key\n- Price: $0.01 / call · Auth: API key · x402: yes · Where: hosted and local\n- Full assessment: https://www.anchorterminal.com/tools/steel.md\n- Against #1: https://www.anchorterminal.com/compare/browser-use-vs-steel.md\n\n### 6. TinyFish, B 67.7/100\n\nTinyFish is a hosted web agent platform from Tiny Fish, Inc. Agents call it through REST APIs for goal-driven automation, remote browser sessions over CDP, search, fetch and research, or through a remote MCP server with OAuth.\n\n- Verdict: OpenAPI documents are published for five APIs, errors carry typed codes and request IDs, and Search and Fetch are free within daily allowances. The terms let Tiny Fish train its models on customer data, run creation takes no idempotency keys, and the status page shows an hour or more of downtime on several days for the Browser and Search APIs.\n- Choose it for: Agents that want one account for a goal-driven web agent, a raw CDP browser, and free search and fetch.\n- Strength: OpenAPI 3.0 documents for the Agent, Browser, Fetch, Search and Research APIs, plus `llms.txt`, `llms-full.txt` and a Markdown twin of every docs page\n- Strength: Search and Fetch are free up to 12,000 requests and 1,000 URLs a day, and new accounts start with $8 of wallet funds with no card\n- Strength: Errors carry a code, a message, optional `details` and a `request_id`. Search and Fetch 429s carry `Retry-After` and `X-RateLimit-*` headers\n- Weakness: The terms of 17 August 2026 allow Tiny Fish to train and improve its models on customer data. No opt-out was found in the reviewed pages\n- Weakness: The status page shows downtime of an hour or more on six days for the Browser API and five for the Search API between 11 July and 8 October 2026, with no incident write-ups\n- Weakness: No idempotency keys on run or session creation. The docs say retrying a failed request may create duplicate runs\n- Price: $0.016 / call · Auth: OAuth or key · x402: no · Where: hosted\n- Full assessment: https://www.anchorterminal.com/tools/tinyfish.md\n- Against #1: https://www.anchorterminal.com/compare/browser-use-vs-tinyfish.md\n\n### 7. Playwright MCP, B 67.6/100\n\nBrowser automation MCP server built on Playwright that drives pages via structured accessibility snapshots rather than screenshots.\n\n- 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.\n- Choose it for: The default choice for an agent that drives web pages by structure rather than pixels. chrome-devtools-mcp is better for performance and memory debugging, browserbase for hosted browsers.\n- Strength: Accessibility snapshots with `depth` limits and `browser_find`, so the agent reads less of the page\n- Strength: 25 tools by default, with network, storage, devtools, vision, PDF and testing behind `--caps`\n- Strength: `readOnlyHint`, `destructiveHint` and `openWorldHint` on every tool\n- Weakness: `browser_run_code_unsafe` runs arbitrary JavaScript in the server process and can't be switched off\n- Weakness: No authentication on the HTTP transport, and `--isolated` is off by default\n- Weakness: Still 0.0.x after 83 releases, with alpha Playwright builds as dependencies\n- Price: Free · OSS · Auth: None · x402: no · Where: local\n- Full assessment: https://www.anchorterminal.com/tools/playwright-mcp.md\n- Against #1: https://www.anchorterminal.com/compare/browser-use-vs-playwright-mcp.md\n\n### 8. Anchor Browser, B 67.3/100\n\nAnchor Browser runs cloud browser sessions for AI agents, controlled through a REST API, a CDP WebSocket, TypeScript and Python SDKs and hosted MCP servers. It adds managed logins with MFA, profiles, proxies, CAPTCHA solving and recordings.\n\n- Verdict: The API has a public OpenAPI 3.1 document, per-unit prices and Agent Access, an unauthenticated flow that issues a trial API key without a dashboard account. No service agreement is published, only a website terms of use, and the documented CDP connection puts the API key in the URL.\n- Choose it for: Agents that must work inside signed-in accounts, with managed identities, MFA handling and a human sign-in link, and agents that need to start with no human signup.\n- Strength: Agent Access issues an API key through two unauthenticated calls and a puzzle, with 1 trial credit, or 5 with a verified-email OIDC token\n- Strength: OpenAPI 3.1 document with 101 paths and 136 operations, llms.txt, llms-full.txt and a Markdown twin of every docs page\n- Strength: Per-unit prices published without login, $0.01 a browser, $0.05 a browser-hour and $8 a GB of proxy traffic\n- Weakness: No service agreement or API terms found. anchorbrowser.io/terms is a website terms of use dated 16 March 2025\n- Weakness: The documented CDP connection passes the API key as `apiKey` in the `wss://connect.anchorbrowser.io` URL\n- Weakness: No changelog, deprecation policy or published SLA found, and docs.anchorbrowser.io/changelog returns 404\n- Price: $0.01 / call · Auth: OAuth or key · x402: no · Where: hosted\n- Full assessment: https://www.anchorterminal.com/tools/anchor-browser.md\n- Against #1: https://www.anchorterminal.com/compare/anchor-browser-vs-browser-use.md\n\n### 9. Hyperbrowser, B 66.9/100\n\nHyperbrowser, from S2 Labs Inc., runs hosted Chromium sessions that agents control over CDP, WebDriver or computer actions, with Fetch, Search, Crawl, Extract and cloud sandbox APIs. Access is by API key, SDKs, a stdio MCP server or x402.\n\n- Verdict: Sessions, Fetch and Search can be bought by x402 at api.hyperbrowser.ai with USDC on Base or Solana and no account, and unit prices are public. The official MCP server was last published in April 2025, no request rate limits or SLA are published, and the privacy notice covers only the website.\n- Choose it for: Agents that need a cloud browser by the minute plus cheap fetch and search, especially wallet-holding agents with no account.\n- Strength: x402 on /x402/session, /x402/web/fetch and /x402/web/search with USDC on Base or Solana and no account, confirmed by a live 402 on 8 October 2026\n- Strength: Unit prices public in the docs, $0.10 a browser-hour billed per second, $0.001 a Fetch page, $0.005 a Search query\n- Strength: Free plan with 5,000 credits and no card, per the pricing page\n- Weakness: The official MCP server's last npm release is 1.0.25 of 13 April 2025, and it isn't in the official MCP registry\n- Weakness: No request rate limits, Retry-After header or SLA found; the pricing page says paid plans have no rate limits\n- Weakness: One `x-api-key` per account with no scopes or rotation guidance found in the reviewed documentation\n- Price: $0.001 / call · Auth: OAuth or key · x402: yes · Where: hosted and local\n- Full assessment: https://www.anchorterminal.com/tools/hyperbrowser.md\n- Against #1: https://www.anchorterminal.com/compare/browser-use-vs-hyperbrowser.md\n\n### 10. agent-browser, B 66.6/100\n\nagent-browser is an open-source command-line tool from Vercel Labs that lets AI agents drive Chrome through accessibility-tree snapshots and element refs. It runs locally as a Rust binary and includes a stdio MCP server.\n\n- Verdict: Snapshots with element refs, a 29-tool default MCP profile and opt-in domain allowlists, action policies and content boundary markers suit agent use. The project is at 0.38.2 with no stable release, the ten most recent CI runs on main failed between 1 and 8 October 2026, and every security control is off by default.\n- Choose it for: Coding agents with a shell that want short commands and compact snapshots.\n- Strength: Accessibility-tree snapshots give each element a ref such as `@e2`, with `-i`, `-c`, `-d` and `--delta` to cut the output\n- Strength: The stdio MCP server loads 29 tools by default and 158 with `--tools all`, each with a typed schema and `readOnlyHint`\n- Strength: Domain allowlist, action policy, action confirmation and content boundary markers are documented with a threat model and known limitations\n- Weakness: The ten most recent CI runs on main, 1 to 8 October 2026, all concluded in failure. In the newest, Windows Rust tests and native end-to-end tests failed\n- Weakness: Version 0.38.2 with a Vercel Labs badge and no stable release or deprecation policy\n- Weakness: All security controls are opt-in. By default there is no limit on navigation, actions or output\n- Price: Free · OSS · Auth: None · x402: no · Where: local\n- Full assessment: https://www.anchorterminal.com/tools/agent-browser.md\n- Against #1: https://www.anchorterminal.com/compare/agent-browser-vs-browser-use.md\n\n4 more are ranked in the full table: https://www.anchorterminal.com/categories/browser.md\n\n## Head to head\n\n- [Browser Use vs Chrome DevTools MCP](https://www.anchorterminal.com/compare/browser-use-vs-chrome-devtools-mcp.md)\n- [Browser Use vs Browserbase](https://www.anchorterminal.com/compare/browser-use-vs-browserbase.md)\n- [Browser Use vs Browserless](https://www.anchorterminal.com/compare/browser-use-vs-browserless.md)\n- [Browser Use vs Steel](https://www.anchorterminal.com/compare/browser-use-vs-steel.md)\n- [Browserbase vs Chrome DevTools MCP](https://www.anchorterminal.com/compare/browserbase-vs-chrome-devtools-mcp.md)\n- [Browserless vs Chrome DevTools MCP](https://www.anchorterminal.com/compare/browserless-vs-chrome-devtools-mcp.md)\n- [Chrome DevTools MCP vs Steel](https://www.anchorterminal.com/compare/chrome-devtools-mcp-vs-steel.md)\n- [Browserbase vs Browserless](https://www.anchorterminal.com/compare/browserbase-vs-browserless.md)\n- [Browserbase vs Steel](https://www.anchorterminal.com/compare/browserbase-vs-steel.md)\n- [Browserless vs Steel](https://www.anchorterminal.com/compare/browserless-vs-steel.md)\n\n## Questions\n\n### What are the highest-rated browser automation tools for AI agents?\n\nBrowser Use has the highest benchmark score of the 14 ranked browser automation tools, 77.9 (BB). Chrome DevTools MCP is second with 77 (BB).\n\n### How many browser automation tools are agent-ready?\n\n5 of the 14 ranked here grade BB or better, the bar for agent-ready on the Anchor benchmark.\n\n### Which browser automation tools accept x402 payments?\n\nBrowser Use, Browserbase, Steel and Hyperbrowser. An agent can pay these per call in USDC with no account.\n\n### Which of these browser automation tools is cheapest?\n\nBy published paid prices, Browser Use, at $0.02 per browser hour, the lowest of the 7 listings here with a paid price in this unit (free allowances aside). Plans, volume tiers and free allowances change the sum, so check the listing's price table.\n\n### How is this list ranked?\n\nBy the Anchor benchmark score out of 100, a weighted mean of the scored categories minus deductions for negative events, from public evidence re-checked as vendors change. Listings cannot pay for a place. The latest assessment behind this page is from 8 October 2026.\n\n## How this list is made\n\nThe order is the Anchor benchmark score, the same number as on each listing. Each listing is graded from public evidence against the benchmark checklist, and the picks are worked out from those grades, prices and facts. No listing pays for its place, and paid audits or listing help never change a score.\n", "meta": { "attribution": "Anchor Terminal (https://www.anchorterminal.com)", "docs": "https://www.anchorterminal.com/docs/", "generatedAt": "2026-10-10", "license": "CC-BY-4.0", "method": "https://www.anchorterminal.com/benchmark/", "methodology": "0.4", "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": "Best of", "url": "https://www.anchorterminal.com/best/" }, { "name": "Browser automation", "url": "" } ], "description": "Browser Use (BB), Chrome DevTools MCP (BB) and Browserbase (BB) lead the 14 ranked browser automation tools. Picks by need, strengths, weaknesses and prices from the Anchor benchmark.", "facts": [ "Browser Use BB", "Chrome DevTools MCP BB", "Browserbase BB" ], "h1": "Best browser automation tools for AI agents", "image": "https://www.anchorterminal.com/assets/og/best-browser.png", "path": "/best/browser/", "published": "", "section": "tools", "title": "Best browser automation tools for AI agents in 2026, ranked", "toc": null, "updated": "2026-10-08", "url": "https://www.anchorterminal.com/best/browser/" }, "tokens": { "markdown": 6300, "slim": 1680 }, "version": 1 }