{"slug": "how-browseract-fixed-the-stale-selector-failures-breaking-my-browser-tasks", "title": "How BrowserAct Fixed the Stale-Selector Failures Breaking My Browser Tasks", "summary": "BrowserAct, a browser automation platform for AI agents, addresses the stale-selector problem that breaks browser tasks when pages re-render with new build hashes. In a test against raw Playwright, BrowserAct's state-based indexing, which uses snapshot-scoped indices instead of selectors, proved resilient to id and class churn. The platform's approach allows agents to reason about the page state turn by turn and detect when the page has changed.", "body_md": "*Disclosure: BrowserAct sponsored this piece. The BrowserAct links below are affiliate-tracked — I get credit if you sign up through them.*\n\nI keep hitting the same failure building agent workflows in Claude Code: the agent captures a selector once, the page re-renders with a new build hash, and the next run breaks on an element that's still right there on the screen — just under a different id. BrowserAct is a browser automation platform built for AI agents — real browser control, persistent browser identity, task sessions, verification handling, and human handoff, all behind one CLI. This test is about one narrow slice of that: how it handles a page whose ids and classes change under you, using Claude Code to drive it.\n\nI tested that against a page I built myself, compared directly against raw Playwright.\n\nHere's the shape of it. A frontend re-renders with fresh CSS-module or styled-components hashes on every deploy. You write:\n\n``` js\nconst id = await page.locator(\"button\").getAttribute(\"id\");\nawait page.reload();\nawait page.locator(`#${id}`).click({ timeout: 3000 });\n```\n\nThat works the day you write it. It breaks the next time the build hash changes, and the failure you get is a bare timeout with no explanation:\n\n```\nlocator.click: Timeout 3000ms exceeded.\n```\n\nPlaywright's own role/text locators fix this specific case — `getByRole(\"button\", { name: \"Submit\" })`\n\nsurvives id/class churn fine, because it never depended on the hash in the first place. That's a real fix, not a workaround. What it doesn't give you is a representation of the page an agent can reason about turn by turn, or a signal that says \"the page under you just changed, stop and re-check.\" That's the gap BrowserAct is actually filling.\n\nBrowserAct doesn't hand Claude Code a selector at all. The loop is: read the current state, choose an action from what's actually there, execute it, reassess.\n\n``` php\nstate -> indexed list of interactive elements, as of right now\nclick -> act on one of them\nstate -> read again, because the page may have changed\n```\n\nThe index is scoped to that one snapshot. It's not a selector, not a stable id — it's a pointer into \"what state just returned,\" and it expires the moment the page does something a new `state`\n\ncall would notice.\n\n```\nuv tool install browser-act-cli --python 3.12\nbrowser-act --version\nbrowser-act browser create --name \"dom-drift-test\" --type chrome --desc \"local churn test\"\nbrowser-act browser list\n```\n\n`browser list`\n\nafter creation is how you get the browser id you'll pass to every session command — it's not returned anywhere else. I used the `chrome`\n\nbrowser type (a local, blank browser, immediate to create) rather than `stealth`\n\n, since this test is about selector churn on a page I control, not anti-bot evasion — `stealth`\n\nrequires an API key and a billed purchase flow that has nothing to do with what I was testing. Creating a local `chrome`\n\nbrowser completed immediately with no purchase page involved.\n\nVersions tested: `browser-act-cli`\n\nv1.1.0, Playwright 1.61.1, Python 3.12.13. To reproduce this: the test page is a small Node server that reshuffles element ids, classes, and order on every reload — no third-party site involved — plus the Playwright scripts run against the same page.\n\nSkill source: [browser-act/skills](https://github.com/browser-act/skills/tree/main/browser-act). Installation details: [docs/installation.md](https://github.com/browser-act/skills/blob/main/docs/installation.md).\n\n```\nbrowser-act --session domtest browser open http://localhost:8934\nbrowser-act --session domtest state\n```\n\n`state`\n\nreturns an indexed list, not a selector:\n\n```\n[1]