Blog
Notes on recording, AI coding agents, and shipping faster.
View the changelog →How I Use ThinkRun to Get the Exact iOS UI I Want From Codex
2026-11-04
My process for mobile UI changes a prompt can't explain. I record the interaction on my iPhone, talk through what I want, and give my coding agent the transcript, screenshots and timestamps.
How I Give Codex a Screen Recording — A Real Narrate Example
2026-10-23
How I share a screen recording with Codex through ThinkRun's transcript, screenshots and timestamps, using a real iPhone app interaction.
Can Claude Code See Images and Screenshots? Yes, Here's How
2026-10-13
Yes, Claude Code can see images. Drag one in, paste it with Ctrl+V (Alt+V on Windows and WSL), or give it a file path. Plus when a recording works better.
How to Give Claude Code a Screen Recording It Can Read
2026-09-24
Turn a screen recording into a transcript, screenshots and timestamps Claude Code can read. Record in your browser, use the extension, or upload a video.
5 Loom alternatives for developers and AI coding agents in 2026
2026-07-29
Compare Loom, ThinkRun, Jam, Cap, Marker.io, and Scribe for async video, developer bug reports, and handoffs to AI coding agents.
Multi-agent browser automation: safe Chrome tab control
2026-07-27
Use local browser automation safely: ThinkRun keeps an AI agent working in its own Chrome tab while protecting tabs owned by other agents.
Chrome DevTools MCP Alternatives (and When Not to Switch)
2026-07-27
Compare Chrome DevTools MCP alternatives, including --slim mode, Playwright MCP, Playwright CLI, browser-use, Browserbase and ThinkRun, and when to stay.
Playwright MCP alternatives — Chrome DevTools MCP, browser-use, and when you want a recording instead
2026-07-25
People searching for a Playwright MCP alternative usually have one of two problems — the tool is burning their context window, or the bug won't reproduce on demand. Those need different answers. An honest comparison of Playwright MCP, Chrome DevTools MCP, browser-use, Browserbase, and recording-based capture.
What 331 days of AI-agent-built software actually looks like (with receipts)
2026-07-17
ThinkRun has been built almost entirely by AI coding agents since August 2025. Here's the honest version of that story — the silent failures, the 13-round review battles, the same-day rollback — and the evidence gallery where we published 40 agent-written versions of our own history.
AI debugging when the agent can't see — how a screen recording became a shipped fix
2026-07-08
Coding agents can drive a browser now — but only for bugs they can reproduce live. We recorded a real production bug with ThinkRun, handed the structured artifact to Codex, and shipped the fix. Here's the whole loop, including the actual artifact the agent read.
I let an AI agent set up Search Console — here's what broke
2026-06-29
I gave an AI agent my real logged-in browser and told it to verify ThinkRun in Google Search Console and Bing. It mostly worked — but the walls it hit are a tour of how the modern web fights automation, and where handing control back to a human is the right call.
How to give Claude Code access to your browser
2026-06-25
Stop pasting screenshots and explaining what went wrong. Record the browser session once and hand Claude Code a structured artifact it can act on — with your real logged-in session.