Browserbase
Also known as: Browserbase MCP
Hosted browsers for AI agents, plus Browserbase Agents, Functions, Fetch and Search on one API key.
Browserbase runs web browsers in the cloud for AI agents and automation. Developers create isolated Chromium sessions, one browser per VM, through an API and drive them with their own agent or with Playwright, Puppeteer or Selenium. The platform handles what breaks agents on the live web: stealth and Verified identity, CAPTCHA solving, proxies, saved contexts for logged in sessions, and regions in the US, EU and Asia.
On top of the browsers, Browserbase sells Browserbase Agents, which take a natural language task, run an autonomous loop with browser, search, shell and file tools in their own session, and return a result. A run can pause when a condition the customer describes arises, such as an approval or a verification code, and resume once they reply. An Optimize tool reviews completed runs and proposes prompt changes, and each Agent tracks completion, failure and timeout rates. Functions host the customer's own browser scripts behind an API, and Fetch and Search give agents web content without a full session.
Every session carries live view, replay and recordings. Organizations use Admin, Contributor and Viewer roles, with SAML single sign on, HIPAA and a DPA on the Scale plan and SOC 2 Type II on every plan. Browserbase also maintains the open source Stagehand SDK, which has its own record in this index and is not graded here. The pages read do not document custom tools for Agents, a memory layer, or a schedule Browserbase runs itself.
Vendor details
Canonical URL
https://www.browserbase.com/
Category
Browser / computer-use agent
Company status
independent
Use cases & customers
Target customers
Deployment options
Integrations
Playwright, Puppeteer, Selenium and CDP on hosted sessions; an MCP server; integrations for Stripe cards, 1Password logins, Box, n8n, Inngest and agent frameworks such as CrewAI, LangChain, Mastra, Agno, Google ADK and Browser Use.
In practice
Your agent needs to operate real websites at scale, but maintaining a fleet of headless browsers with proxies and anti-bot handling is eating your team. Browserbase runs managed Chromium sessions through one API so you don't have to.
Your Playwright scripts keep breaking when sites change their markup. Browserbase's open-source Stagehand SDK lets you write actions as plain-English instructions that an AI resolves at runtime and self-heal across page redesigns.
Your agent keeps hitting CAPTCHAs and login walls, and you can't see why runs fail. Browserbase adds stealth fingerprints, Agent Identity to get past bot detection, and session recordings and a live view for debugging.
Sources & related URLs
Research sources
Agentic Index coverage score
10.0 / 14 capabilities · 71%
| Integrations & Tool Calling | Partial |
|---|---|
|
Browserbase Agents have four built in tool groups (browser, search, shell and files) and the docs state they cannot take custom tools yet; the integrations directory (Stripe cards, 1Password logins, Box, n8n and agent frameworks) connects services to sessions the customer's own code drives. Sourcedocs.browserbase.com/platform/agents/how-it-worksread 2026-09-27 |
|
| Workflow Orchestration | Partial |
|
An Agent run plans and executes multi step browser work, reusable Agents run one prompt across many inputs, and Functions host the customer's own scripts behind an API; sequencing, branching and retries across runs live in the customer's code, and no workflow model is documented. Sourcedocs.browserbase.com/platform/functions/overviewread 2026-09-27 |
|
| Knowledge Grounding & RAG | Not documented |
|
No retrieval over the customer's own knowledge is documented; Search and Fetch read the public web at run time. Sourcedocs.browserbase.com/llms.txtread 2026-09-27 |
|
| Human Oversight & Guardrails | Full |
|
A run given pauseWhen pauses when the stated condition arises, such as an approval or a verification code, tells the customer what it needs and continues where it left off once they reply; Allowed Domains limits where sessions may browse. Sourcedocs.browserbase.com/platform/agents/pausing-runsread 2026-09-27 |
|
| Security, Identity & Governance | Full |
|
Organizations assign Admin, Contributor and Viewer roles and can use SAML single sign on on the Scale plan; SOC 2 Type II on every plan, HIPAA with a BAA, a DPA and third party penetration tests, with a trust center at trust.browserbase.com. Sourcedocs.browserbase.com/account/enterprise/securityread 2026-09-27 |
|
| Observability & Auditability | Full |
|
Each Agent run keeps a chronological message transcript of what the Agent did, and every session has live view, session replay and downloadable recordings, with logs retained 7 to 30 or more days by plan and optional to disable. Sourcedocs.browserbase.com/platform/browser/observability/observabilityread 2026-09-27 |
|
| Memory & State Persistence | Partial |
|
An Agent reads its memory of prior steps within a run; Contexts persist cookies, local storage and cache across sessions, which is saved browser state rather than a memory layer, and no store the Agent reads across runs is documented. Sourcedocs.browserbase.com/platform/browser/core-features/contextsread 2026-09-27 |
|
| Deployment & Data Residency | Full |
|
The Sessions API takes a region parameter and data residency is configurable across US, EU and Asia regions, with zero data retention and bring your own storage for enterprise accounts; the service is hosted only. Sourcedocs.browserbase.com/optimizations/latency/multi-regionread 2026-09-27 |
|
| Prebuilt Agents, Templates & Packs | Not documented |
|
The Agents examples are worked task recipes (KYC checks, price monitoring, many portals) the customer writes into their own Agent, and Skills teach coding agents to use Browserbase; no packaged agents or templates to adopt are documented. Sourcedocs.browserbase.com/platform/agents/examplesread 2026-09-27 |
|
| Triggers & Channel Coverage | Partial |
|
Runs and Functions start from the API, SDKs, the Browse CLI or the MCP server, and the docs point to the customer's own webhook, queue worker or scheduled job to invoke a Function on a cron; Browserbase's webhooks are outbound notices, and no schedule Browserbase runs itself is documented. Sourcedocs.browserbase.com/platform/webhooks/overviewread 2026-09-27 |
|
| Model Flexibility & Routing | Full |
|
Model Gateway lets a customer pick OpenAI, Anthropic or Google models by name on a Browserbase API key, and enterprise security lists bring your own LLM; Browserbase Agents themselves take no model setting, and the gateway is consumed through the Stagehand SDK. Sourcedocs.browserbase.com/platform/model-gateway/overviewread 2026-09-27 |
|
| APIs, SDKs & MCP Extensibility | Full |
|
A documented REST API for sessions, contexts, Agents, Functions, Fetch and Search with Node.js and Python SDKs, the Browse CLI and a Browserbase MCP server. Sourcedocs.browserbase.com/reference/introductionread 2026-09-27 |
|
| Testing, Debugging & Optimization | Full |
|
The Optimize tool analyzes completed Agent runs and proposes a system prompt change to re run, and each Agent tracks completion, fail and timeout rates and average duration over its run history, an optimization loop on the deployed Agent. Sourcedocs.browserbase.com/platform/agents/optimizing-agentsread 2026-09-27 |
|
| Browser & Computer Use | Full |
|
Browserbase runs hosted Chromium browsers, one per VM, that agents and Playwright, Puppeteer or Selenium code drive, with stealth, CAPTCHA solving, proxies and live view, and Browserbase Agents browse, click, type and extract on the web in their own session. Sourcedocs.browserbase.com/platform/agents/overviewread 2026-09-27 |
|
The Agentic Index coverage score grades every vendor Full, Partial or Not documented against the same 14 buyer facing capabilities, from public evidence only. Each capability links to how all vendors in the index score on it. How this evidence is graded
Recent platform changes
Secrets are now generally available in Browserbase Functions, so credentials and API keys are stored in Browserbase, attached to specific functions and read at run time instead of being written into code. Values are encrypted on the client and from listings.
Bears on: Security / enterprise
View sourceBrowserbase Agents can pause a run for human input and resume within the same browser session. Developers set a natural language pauseWhen condition, which gives the agent a pause tool and lets it enter PAUSED when the condition is met. The resume API accepts a follow up task, such as continuing after a user supplies an MFA code.
Bears on: Browser/computer use
View sourceBrowserbase made Search, the feature that gives agents read access to the internet, natively available inside the dashboard. Users can sample and test search queries directly in the UI before deploying them to production.
Bears on: Agent capability
View sourcePricing
From $20 a month (Developer) · free tier
hybrid
Included quota
Free ($0): 1 browser hour, no proxy, 3 concurrent, 15 minute sessions, 3 Agent calls, 1,000 Search and 1,000 Fetch calls, 7 day retention, 1 project. Developer ($20): 100 browser hours then $0.12/hr, 1 GB proxy then $12/GB, 25 concurrent, 6 hour sessions, 15 Agent calls, basic Verified identity and auto CAPTCHA, 7 day retention, 2 projects. Startup ($99): 500 hours then $0.10/hr, 5 GB proxy then $10/GB, 100 concurrent, 50 Agent calls, 10,000 Fetch calls, 30 day retention, 5 projects, priority support. Scale (custom): 250+ concurrent, Verified identity, 30+ day retention, SSO, HIPAA BAA and DPA. Functions are free on every plan.
What is public
Browserbase publishes four plans: Free, Developer at $20 a month, Startup at $99 a month and a custom Scale plan, each with browser hours, proxy bandwidth and Agent, Search and Fetch call allocations and published overage rates. SOC 2 covers every plan; HIPAA, a DPA and SSO are on Scale.
Billing mechanics
Monthly plan fee with allocations of browser hours, proxy GB and API calls; usage past an allocation bills at the plan's overage rate with no cut off. Browser time bills by the minute and proxy by the MB, with a one minute and one MB minimum per session. Hosted Functions are free.
Cost watchouts
Proxy traffic at $10 to $12 per GB past the allocation; Fetch and Extract cost more through proxies; retention is 7 days below Startup.
Variable cost rationale
Hybrid and genuinely usage-driven - a flat tier fee plus metered browser hours, proxy GB, and compute GB-seconds; cost scales with session length, concurrency, and proxy use, so heavy/long automation runs can far exceed the base subscription
Additional watchouts
Several meters (browser hours, proxy GB, Agent, Search, Fetch and Extract calls) sit beside the plan fee; the Free plan's 15 minute sessions and one hour are for testing; SSO, HIPAA and Verified identity need the Scale plan.
Overage / add-ons
Allocations, not limits: browser hours at $0.12/hr (Developer) or $0.10/hr (Startup), proxy at $12 or $10 per GB, Search $7 per 1,000 calls, Fetch $1 or $0.50 per 1,000, Extract $4 per 1,000 (more with proxies); Scale is custom.
Sales call required
Mixed (some tiers require a call)
Free / trial
Free plan, no card: 1 browser hour, 3 concurrent sessions, 3 Agent calls
Lowest paid plan
Developer, $20 a month
Commercial notes
Browserbase hosts browsers for AI agents and sells Browserbase Agents, Functions, Fetch, Search and a Model Gateway on the same plans; it also maintains the open source Stagehand SDK, which has its own record in this index.
Key ambiguities
Agent call overage and Scale terms are not published.
Cancellation / refund
Free plan without a card; Developer and Startup monthly with overage; Scale by contract.
Support SLA / resale
Email (Free), standard (Developer), priority (Startup) and high priority support (Scale).
Missing data
Scale plan pricing and Agent call overage rates are not published.
Related vendors
- AgentQL — A self-healing query language for the web: natural language…
- AGI, Inc. — AGI, Inc
- Airtop — Cloud browsers for AI agents that compile plain English workflows…
- Asteroid — Healthcare portal integration platform: supervised browser and…
- Automat AI — AI agents that operate computers visually, sold as Automat Workforce…
- Autotab — General AI agent that learns a workflow from a demonstration and…
Alternatives to Browserbase
The closest documented capability profiles to Browserbase among browser and computer-use agents tracked by Agentic Index, ordered by similarity on the same 14 point evidence the rankings use. No vendor pays for placement.
- Browser Use11.5 / 14Adds documented Prebuilt Agents, Templates & PacksBrowserbase vs Browser Use →
- Notte9.5 / 14Fuller documented coverage on Workflow Orchestration and Triggers & Channel Coverage
- Skyvern11.5 / 14Adds documented Prebuilt Agents, Templates & Packs
- Island9.0 / 14Adds documented Knowledge Grounding & RAGBrowserbase vs Island →
- Simular10.0 / 14Adds documented Prebuilt Agents, Templates & Packs
- Automat AI11.5 / 14Adds documented Knowledge Grounding & RAG and Prebuilt Agents, Templates & Packs
Similarity is computed from each vendor's Agentic Index coverage score evidence, axis by axis, not from the totals. How this evidence is graded