AI News HubLIVE
站内改写6 min read

Launch HN: Intuned (YC S22) – Build and run reliable browser automations as code

Intuned is a platform that lets you describe what you need in natural language, and it generates production-ready Playwright code, deploys it, and fixes it when sites change. It supports scrapers, crawlers, RPA, AI automation, and managed scraping. Key features include stealth mode, captcha solving, authentication handling, scheduling, auto-scaling, and a Web Tasks API that learns from reuse.

SourceHacker News AIAuthor: fkilaiwi

Stop writing browser automation

Describe what you need. Intuned Agent builds production-ready Playwright code, deploys it, and fixes it when sites change. You get the reliability of code without writing it yourself.

Start for freeView docs

Questions? , Talk to a human or

Something Else

USECASE

Scrapers

Extract data from any website—e-commerce, government portals, job boards, and more. Use Intuned Agent to generate and maintain your Playwright code. Deploy with built-in stealth, auth, scheduling, and scale.

Key Features:

Intuned Agent — Generate scrapers from a prompt and schema, fix issues with AI when sites change

Your code, your way — Write in TypeScript or Python using Playwright or any compatible library or framework

Stealth & auth — Anti-detection, captcha solving, login handling built in

Scheduled jobs & monitoring — Recurring runs with full logs and session recordings

Auto-scaling — One machine or hundreds, we handle it

Try Intuned AgentView Examples

USECASE

Crawlers

Discover, navigate, and collect data across thousands of pages. Build crawlers that follow links, parse sitemaps, and index content—with built-in stealth, scheduling, and scale.

Key Features:

Your code, your way — Write in TypeScript or Python using Playwright or any compatible crawling library or framework

Crawl4AI support — First-class support for Crawl4AI—deploy and scale your crawlers without managing infrastructure

Stealth mode — Anti-detection, proxies, and captcha solving built in

Scheduled jobs & monitoring — Recurring crawls with full logs and session recordings

Scalable infrastructure — Control how many machines to run on, we handle the rest

Start for FreeView Examples

USECASE

RPA

Build browser automations that interact with websites on behalf of your users—form submissions, data entry, account actions, and more. Deploy with built-in auth handling, stealth, and concurrency control. Perfect for integrating with services that don't offer APIs.

Key Features:

Built-in authentication support — Write create.ts and check.ts (or .py), Intuned manages the session lifecycle—validating, reusing, and recreating as needed

Your code, your way — TypeScript or Python with Playwright, or AI-driven via Anthropic Computer Use, OpenAI CUA, Stagehand, Browser-use, or Gemini Computer Use

UI actions or network calls — Automate via browser interactions or faster in-browser network requests

Stealth mode — Anti-detection, proxies, and captcha-solving built in

API-first with concurrency control — Deploy as an API endpoint, set parallel limits, we queue and execute

Start for FreeView Examples

USECASE

AI Automation

Build AI-powered automations that can navigate, interact, and extract from any website. Whether you're building AI agents as a product or using AI to automate scraping and workflows—deploy with built-in stealth, auth, and scale.

Key Features:

Works with any AI browser library or API — Anthropic Computer Use, OpenAI CUA, Stagehand, Browser-use, Gemini Computer Use, Intuned/Browser SDKs, and more

Mix code and AI — Combine deterministic Playwright logic with AI-driven steps in a single automation

Stealth mode — Anti-detection, proxies, and captcha solving built in

Built-in authentication support — Write create.ts and check.ts (or .py), Intuned manages the session lifecycle - you can use AI for that too

AI usage controls — Monitor and limit AI credit usage across your automations

Scalable infrastructure — Deploy as an API or schedule jobs, control concurrency, we handle the rest

Start for FreeView Examples

USECASE

Managed Scraping

Need hundreds or thousands of scrapers? Our solution engineers—powered by Intuned Agent—build and maintain them for you. We've delivered thousands of production scrapers. You tell us what data you need, we handle the rest.

Key Features:

We build, you ship — Tell us the sites and schemas, we deliver production-ready scrapers

Cost-effective delivery — Solution engineers + Intuned Agent means faster builds at lower cost

Ongoing maintenance — Sites change, we update your scrapers so they keep running

Your code, no lock-in — You own the code, switch to self-serve anytime, no reselling or black boxes

Flexible pricing — Flat fee per source or custom arrangements

Built on Intuned infrastructure — Same stealth, auth, scale, and monitoring you'd get building yourself

Talk to Us

USECASE

Something Else

Have a unique use case? We support custom browser automation solutions tailored to your specific needs.

Usecases

Scrapers

Extract data from any website—e-commerce, government portals, job boards, and more. Use Intuned Agent to generate and maintain your Playwright code. Deploy with built-in stealth, auth, scheduling, and scale.

Key Features:

Intuned Agent — Generate scrapers from a prompt and schema, fix issues with AI when sites change

Your code, your way — Write in TypeScript or Python using Playwright or any compatible library or framework

Stealth & auth — Anti-detection, captcha solving, login handling built in

Scheduled jobs & monitoring — Recurring runs with full logs and session recordings

Auto-scaling — One machine or hundreds, we handle it

Try Intuned AgentView Examples

Crawlers

Discover, navigate, and collect data across thousands of pages. Build crawlers that follow links, parse sitemaps, and index content—with built-in stealth, scheduling, and scale.

Key Features:

Your code, your way — Write in TypeScript or Python using Playwright or any compatible crawling library or framework

Crawl4AI support — First-class support for Crawl4AI—deploy and scale your crawlers without managing infrastructure

Stealth mode — Anti-detection, proxies, and captcha solving built in

Scheduled jobs & monitoring — Recurring crawls with full logs and session recordings

Scalable infrastructure — Control how many machines to run on, we handle the rest

Start for FreeView Examples

RPA

Build browser automations that interact with websites on behalf of your users—form submissions, data entry, account actions, and more. Deploy with built-in auth handling, stealth, and concurrency control. Perfect for integrating with services that don't offer APIs.

Key Features:

Built-in authentication support — Write create.ts and check.ts (or .py), Intuned manages the session lifecycle—validating, reusing, and recreating as needed

Your code, your way — TypeScript or Python with Playwright, or AI-driven via Anthropic Computer Use, OpenAI CUA, Stagehand, Browser-use, or Gemini Computer Use

UI actions or network calls — Automate via browser interactions or faster in-browser network requests

Stealth mode — Anti-detection, proxies, and captcha-solving built in

API-first with concurrency control — Deploy as an API endpoint, set parallel limits, we queue and execute

Start for FreeView Examples

AI Automation

Build AI-powered automations that can navigate, interact, and extract from any website. Whether you're building AI agents as a product or using AI to automate scraping and workflows—deploy with built-in stealth, auth, and scale.

Key Features:

Works with any AI browser library or API — Anthropic Computer Use, OpenAI CUA, Stagehand, Browser-use, Gemini Computer Use, Intuned/Browser SDKs, and more

Mix code and AI — Combine deterministic Playwright logic with AI-driven steps in a single automation

Stealth mode — Anti-detection, proxies, and captcha solving built in

Built-in authentication support — Write create.ts and check.ts (or .py), Intuned manages the session lifecycle - you can use AI for that too

AI usage controls — Monitor and limit AI credit usage across your automations

Scalable infrastructure — Deploy as an API or schedule jobs, control concurrency, we handle the rest

Start for FreeView Examples

Managed Scraping

Need hundreds or thousands of scrapers? Our solution engineers—powered by Intuned Agent—build and maintain them for you. We've delivered thousands of production scrapers. You tell us what data you need, we handle the rest.

Key Features:

We build, you ship — Tell us the sites and schemas, we deliver production-ready scrapers

Cost-effective delivery — Solution engineers + Intuned Agent means faster builds at lower cost

Ongoing maintenance — Sites change, we update your scrapers so they keep running

Your code, no lock-in — You own the code, switch to self-serve anytime, no reselling or black boxes

Flexible pricing — Flat fee per source or custom arrangements

Built on Intuned infrastructure — Same stealth, auth, scale, and monitoring you'd get building yourself

Talk to Us

Looking for something else?

Web Tasks API

Web Tasks that learn each run

Run any web task from a prompt. Add a reuseKey and it learns —faster and cheaper with every run.

const companySchema = z.array( z.object({ name: z.string(), location: z.string(), }), );

await webTasks.run({ task: "Scrape YC companies. Return name, batch, description, and URL.", startUrl: "https://www.ycombinator.com/companies", parameters: { batch: "S24" }, outputSchema: companySchema, reuseKey: "yc_companies_scraper", });

// ✓ created skill yc_companies_scraper // ✓ wrote reusable helpers // ✓ returned 247 companies · 6m 12s · $0.84 AI + 0.10 compute hrs

await webTasks.run({ task: "Scrape YC companies. Return name, batch, description, and URL.", startUrl: "https://www.ycombinator.com/companies", parameters: { batch: "W24" }, outputSchema: companySchema, reuseKey: "yc_companies_scraper", });

// ✓ reused skill yc_companies_scraper // ✓ returned 208 companies · 58s · $0.12 AI + 0.02 compute hrs

class Company(BaseModel): name: str location: str

await web_tasks.run( task="Scrape YC companies. Return name, batch, description, and URL.", start_url="https://www.ycombinator.com/companies", parameters={"batch": "S24"}, output_schema=list[Company], reuse_key="yc_companies_scraper", )

✓ created skill yc_companies_scraper

✓ wrote reusable helpers

✓ returned 247 companies · 6m 12s · $0.84 AI + 0.10 compute hrs

await web_tasks.run( task="Scrape YC companies. Return name, batch, description, and URL.", start_url="https://www.ycombinator.com/companies", parameters={"batch": "W24"}, output_schema=list[Company], reuse_key="yc_companies_scraper", )

✓ reused skill yc_companies_scraper

✓ returned 208 companies · 58s · $0.12 AI + 0.02 compute hrs

Intuned Web Tasks runs browser automations from natural-language instructions — scraping, crawling, form filling, multi-step actions, and downloading files. Use it whenever the user asks to extract data from, act on, or pull files from a website.

Flow

If no task has been given yet, ask what they want to automate in one line and offer 1–2 examples — e.g. "What would you like to automate? For example, 'scrape all companies from the YC directory' or 'download every PDF from a site'."

When the user gives you an automation task:

  1. Ask whether to trigger via the Intuned CLI or the SDK (TypeScript or Python).
  2. Trigger it, using their request as the task. Don't ask anything else

unless a required input is missing.

Optional inputs

Include these only when they apply and you can infer them from the task — never ask for them:

  • parameters — dynamic inputs that change between runs.
  • outputSchema — JSON Schema, only when extracting structured data.
  • startUrl — only when known; the agent infers it otherwise.
  • reuseKey — when you re-run the same or similar task (e.g. to change a limit or

parameter), reuse the reuse.key the previous run returned, or a key the user gives you. Only omit it for a brand-new task; the server then generates one — keep it and reuse it on follow-ups for the same/similar task. Never ask whether the task is repetitive.

CLI

Install first if missing: npm install -g @Intuned/cli, verify with intuned -v. intuned platform webtasks run "" --json Add --reuse-key when re-running the same/similar task (the prior run's key) or when the user provided one.

TypeScr

[truncated for AI cost control]