ego(lite) articles
Everything you need to know about browser automation
Guides, insights, comparisons, and reviews across Playwright MCP, Browser Use, Chrome DevTools MCP, coding agents, and the browsers they drive.

Guides
How to keep browser automations stable long term
A browser script can keep running after its result has stopped being trustworthy. Build a maintenance loop around accepted outcomes, change checks, safe recovery and clear ownership.
Nov 8, 2026 · 9 min read

Guides
Fixing scraper breakage when websites change
A scraper can return zero rows and still report success. Four page versions show how extraction layers and field checks catch that failure.
Nov 8, 2026 · 12 min read

Guides
Agent harness: what it is, how to build one, how to choose
Agent equals model plus harness. Three authorities currently disagree about where a harness ends and a runtime begins, so this starts by settling that, then builds a small one, then looks at the part almost nobody writes about: what the agent actually sees on the web.
Nov 7, 2026 · 18 min read

Guides
Codex Chrome extension: setup, permissions, and fixes
The official browser-extension path used by Codex, with current setup, supported browsers, permissions, troubleshooting, developer access, and alternatives.
Nov 7, 2026 · 16 min read

Guides
Google Antigravity browser automation: setup and fixes
Google Antigravity browser automation: setup, fixes, and alternatives
Nov 7, 2026 · 16 min read
Guides
agent-browser video recording: capture and verify
A video preserves the browser's visible steps. Our local Request Desk run shows how to pair that recording with a fresh page and server record before calling the task done.
Nov 2, 2026 · 15 min read
Guides
Claude Agent SDK browser automation: verify the write
Our custom MCP browser tool failed twice before submission. After an adapter fix, three local tasks stored a Review and three HTTP 503 tasks stopped without a write.
Nov 2, 2026 · 11 min read

Guides
Playwright authentication: login once, reuse state
Playwright authentication fails in three recognizable ways, and two of them look like bugs elsewhere: you land back on the login page, an API returns 401, or a post-login element never appears. This guide separates the four root causes, shows the login-once flow with real measured runs, and covers expiry detection, automatic re-authentication, parallel account isolation, and the verification steps that prove the session is genuinely live before you rely on it.
Nov 2, 2026 · 16 min read

Guides
Chrome DevTools MCP: setup, existing sessions, and fixes
Install Chrome DevTools MCP, connect an agent to an existing Chrome session, and diagnose the Node, profile, permission, and privacy failures that stop first runs.
Oct 31, 2026 · 18 min read