Browser Tools SDK

Browser Tools SDK

Browser Tools SDK is an open-source npm package from Libretto that gives AI agents a small set of Playwright-backed browser tools so agents can open real browsers, read page snapshots, and run Playwright actions programmatically.

Browser Tools SDK is automation software teams evaluate for automation. Use this page to review pricing, integration signals, and the best alternatives before you commit.

Contact for pricing
#103 in Automation (103 tools)
Just launched
Data reviewed Jul 22, 2026

Quick Overview

Best for: Automation

What it does

Automation software for decision-makers comparing workflow fit and alternatives.

Best fit

Automation

Pricing snapshot

Contact for pricing

Next step

Compare Browser Tools SDK with similar tools before you shortlist it.

Compare this tool before you shortlist it

Review alternatives, pricing posture, and workflow fit side by side.

Browser Tools SDK

Browser Tools SDK is an open-source SDK from Libretto that provides six small tools enabling AI agents to operate real browsers via Playwright. The SDK exposes browser-level primitives (open, snapshot, exec, status, close, connect/inspect) so agents can read a compact accessibility-style snapshot of a page and then run Playwright code to interact with the live page. It is distributed as an npm package (libretto-browser-tools) and includes adapters for common agent frameworks such as AI SDK and Pi, plus support for local or hosted browser providers.

Six tools let any AI agent open a real browser, read the page, and act with Playwright.

Own this listing?

Claim this page to add pricing, features, screenshots, and verified owner details.

Claim this listing

Key Features

Six browser tools

A compact toolset (browser_open, browser_snapshot, browser_exec, browser_status, browser_close and inspect/connect) that lets agents read page structure and run Playwright actions.

browser_snapshot

Returns a compact accessibility tree with stable refs so the agent can read page structure without raw HTML.

browser_exec

Runs Playwright code on the live page; each call returns a compact snapshot diff showing what changed.

Built-in adapters

Includes adapters for AI SDK and Pi and provides a custom path to connect other agent frameworks.

Local and hosted browser providers

Supports running Chromium locally with LocalBrowserProvider and built-in providers including Libretto Cloud, Browserbase, Kernel, and Steel.

Open source distribution

The package is available under the MIT license in the Libretto repository.

Libretto CLI & tooling

Integrates with Libretto CLI to turn recorded browser workflows into reusable TypeScript APIs and to debug or repair failed browser runs.

Benchmarked efficiency

Published benchmarks on the page claim lower outcome cost vs alternatives (example: Browser Tools $0.106 vs dev-browser $0.257 in a 26-task measurement).

Pricing

Claim this listing to add current pricing tiers.

Use Cases

Give agents real browser access

Allow AI agents to open a real browser, read a compact snapshot of the page, and execute Playwright steps to complete tasks on live websites.

Automate complex page interactions

Enable agents to work through complex pages by combining browser_snapshot for reading structure and browser_exec for interacting with the live DOM.

Record and expose browser workflows as APIs

Use Libretto CLI to record live workflows and save them as reusable TypeScript APIs.

Debug and repair failed browser runs

Give a failed run to an agent and receive a tested pull request or repaired workflow using Libretto's debugging tools.

Integrations

AI SDK adapter

Built-in adapter to connect the Browser Tools SDK to the AI SDK agent framework.

Pi adapter

Built-in adapter to connect the Browser Tools SDK to Pi.

Custom adapter

A custom integration path for wiring the base tools into other agent loops or frameworks.

LocalBrowserProvider

Run Chromium on your machine as a browser provider.

Hosted providers (Libretto Cloud, Browserbase, Kernel, Steel)

Built-in providers listed on the page for running browsers in hosted environments.

Benefits

Lets AI agents read page structure without raw HTML by returning compact accessibility-style snapshots.
Enables agents to run Playwright code on live pages and receive snapshot diffs of what changed.
Includes adapters for common agent frameworks, so you can keep your current model and agent loop.
Open-source MIT-licensed package available via npm for developer adoption.
Benchmarked to reduce outcome cost compared to several alternatives on public website tasks.

Limitations

Browser Tools SDK does not replace Playwright; it runs agent Playwright code through Playwright and you retain the Playwright API.
The SDK intentionally exposes only six tools; more complex or unconventional browser interactions may still require direct Playwright use or custom tooling.

Frequently Asked Questions

Which agent frameworks work today?
The package includes adapters for AI SDK and Pi. The base tools are framework-neutral, so you can also wire them into your own agent loop.
Which browser providers can I use?
Run Chromium on your machine with LocalBrowserProvider, or use the built-in providers for Libretto Cloud, Browserbase, Kernel, and Steel.
Why does the SDK expose only six tools?
Most browser work needs two tools: browser_snapshot to read the page and browser_exec to run Playwright. The other four tools open, connect, inspect, and close browser sessions.
Does Browser Tools SDK replace Playwright?
No. It gives an agent a small tool set and runs its browser_exec code through Playwright. You still have the Playwright API.
Is Browser Tools SDK open source?
Yes. The package is available under the MIT license in the Libretto repository.

Getting Started

  1. 1 Install the package: $npm i libretto-browser-tools
  2. 2 Create or select a browser provider (for example: const provider = new LocalBrowserProvider()).
  3. 3 Create the SDK tools and use them with your agent: const { tools } = createAiSdkBrowserTools(provider); then call your agent/generation API with the tools (example shown: generateText with model, tools, prompt).
  4. 4 Optionally use Libretto CLI to record workflows, turn them into reusable TypeScript files, and debug failed runs.

Support

docs

Documentation and usage examples available on the Libretto site (Docs link referenced on the page).

forum

Libretto forum is referenced on the site for community discussion.

cli & repo

VIEW SOURCE ON GITHUB and Libretto CLI tooling are referenced on the page for debugging and workflow tooling.

blog & changelog

Blog and changelog links are available from the site for updates and release notes.

API

Available: No

Compare Browser Tools SDK with similar tools

See how it stacks up against alternatives

Related Tools

View all 103 →
Contact for pricing
nventr

nventr

nventr is an enterprise AI platform that orchestrates models, agents, and business systems into scalable workflows for automation, intelligent document processing, and real-time decisioning across industries.

Automation
High-growth
Free
Webfill

Webfill

WebFill is a browser extension that uses advanced AI models to automate form filling, surveys, multiple-choice questions, email generation and response, and general data-entry tasks for students and professionals.

Automation
Paid
Docgpt

Docgpt

GPT for Sheets (by DocGPT.ai) is a Google Sheets add-on that embeds 100+ AI models and spreadsheet AI formulas to perform lead enrichment, web scraping, SEO content generation, image/vision tasks, automation and bulk workflows directly inside Google Sheets.

Automation
Enterprise-ready High-growth
Freemium
Parsio

Parsio

Parsio is an AI-powered document and email parser that automatically extracts structured data from PDFs, emails and other documents using AI, GPT, template-based parsing and OCR, intended for business automation and integrations.

Automation
High-growth
Paid
Naratix

Naratix

Naratix provides enterprise-grade ecommerce automation and catalog intelligence to automate product data enrichment, content and image creation, price monitoring, taxonomy and category management, and multi-channel publishing at scale.

Automation
Enterprise-ready
Freemium
Involve

Involve

involve.me is an AI-powered quiz funnel builder that combines interactive quizzes, calculators, forms and landing pages with built-in email automation, CRM features, A/B testing, analytics and payments to capture, qualify and convert leads.

Automation
Free
Timelines

Timelines

TimelinesAI is a WhatsApp CRM integration and unified team inbox that syncs WhatsApp conversations to CRMs, enables shared multi-number management, and automates workflows with AI-powered automation and a visual workflow builder.

Automation
Paid
Process

Process

Process Street is a compliance operations platform that centralizes policies and standard operating procedures, automates workflows with AI, and delivers audit-ready proof to help organizations standardize and enforce processes across teams and industries.

Automation
Enterprise-ready

Premium Alternatives

Paid
OTP Inspired actor supervisor based full stack templates

OTP Inspired actor supervisor based full stack templates

ShipStacks provides production-grade, OTP-inspired full-stack SaaS templates that include supervisors/actor patterns, auth, payments, uploads, AI chat and agent playbooks, and Docker-ready deployment in multiple languages and frameworks.

Developer Tools
High-growth
Paid
ClaudeThings

ClaudeThings

ClaudeThings provides a packaged, continuously-updating set of 89 specialized agents, 103 pre-built skills, and 181 slash commands that act as an AI engineering and marketing team for Claude Code — delivered as a private GitHub repo and installed with a single npx command. It adapts to any stack via a CLAUDE.md project manifest and is sold as a one-time purchase with lifetime updates.

AI Agents
High-growth
Paid
hybridai

hybridai

HybridAI (HybridClaw) provides enterprise-grade, EU-hosted AI agents and a central control layer to build, deploy, and manage custom AI coworkers for VAT automation, BI, compliance, and other business workflows, with GDPR & AI Act compliance.

AI Agents
High-growth
Paid
veggie-ai

veggie-ai

Veggie AI is a web-based tool that uses AI to generate controllable short videos from uploaded character photos, action videos, or text prompts, offering multiple creation modes (Mix, Animate, Ideate, Stylize) and downloadable outputs.

Video Generation
High-growth
Paid
Aiportraitgen

Aiportraitgen

AI Portrait Gen is an online AI portrait generator that creates realistic, high-quality portrait photos from a few user-supplied photos. It offers customizable locations, outfits and styles, pay-as-you-go credits (no subscription), and privacy protections for uploaded images.

Image & Design
Paid
Kling3

Kling3

Kling 3 is an AI-powered video and image generation platform (Kuaishou's third-generation model) that creates cinematic 4K images and up to 15-second videos with character consistency, multilingual lip-sync, and integrated audio.

Video Generation
Paid
shorts-faceless

shorts-faceless

ShortsFaceless is an AI-powered platform that automates creation of faceless short-form videos (YouTube Shorts, TikTok, Reels) by generating scripts, images, voiceovers, subtitles and exporting HD videos to help creators scale production quickly.

Video Generation
High-growth
Paid
Documentpro

Documentpro

DocumentPro is an API-first document intelligence platform that extracts structured data from invoices, purchase orders, tax forms and other documents for embedding into software platforms; it emphasizes fast integration, AI-powered extraction, and maintenance-free production operation.

Business Intelligence
Enterprise-ready

Explore Related Categories

Explore by Outcome