Record real iPhone sessions and let a Claude Code skill find the bugs
ios-black-box works like a flight recorder for your iOS app: it records the screen, what the app says out loud and an event log of everything it did on a real iPhone, and a Claude Code skill reads the recording back and finds the bugs.
Source
@aydin_ayanzadeh on X · 2026-10-18
Written in our own words. Credit to the original author. Link checked 2026-10-19.
What they did
ios-black-box works like a flight recorder for your iOS app: it records the screen, what the app says out loud and an event log of everything it did on a real iPhone, and a Claude Code skill reads the recording back and finds the bugs.
Result (as claimed)
Open source; the builder's point is that the simulator guesses while the real phone shows what happened.
Tools they used
Cost: Claude Code: from $17/mo.
Set it up
Pick a contained repo or task
Start on a branch or side project where mistakes are cheap.
Install and configure
Set up Claude Code. Read what it can run on your machine and limit permissions where possible.
Run it on a real task
Give it a well-scoped issue and let it work. Review the diff like any pull request.
Add guardrails
Keep tests and CI in the loop, and never let an agent push to main or touch production secrets unattended.
Copy the full setup prompt
Paste into your agent and it will walk you through the setup, asking before anything that sends, posts or spends.
You are helping me set up this AI use case end to end in Claude Code. Goal: Record real iPhone sessions and let a Claude Code skill find the bugs What the original author did: ios-black-box works like a flight recorder for your iOS app: it records the screen, what the app says out loud and an event log of everything it did on a real iPhone, and a Claude Code skill reads the recording back and finds the bugs. Reported result (as claimed by the author): Open source; the builder's point is that the simulator guesses while the real phone shows what happened. Source: https://x.com/aydin_ayanzadeh/status/2111674935554777493 Tools they used: Claude Code Category: Dev & ops Please: 1. Ask me 3 short questions about my situation (accounts I have, budget, what I want the output to look like). 2. Propose a step-by-step plan using these tools or cheaper alternatives, with the time and cost of each step. Use official pricing pages; say "check pricing" if unsure. 3. Then do the setup with me one step at a time, waiting for my OK before anything that sends messages, posts publicly, spends money or deletes data. 4. Finish with a checklist to run it again next time.
You are helping me set up this AI use case end to end in Codex. Goal: Record real iPhone sessions and let a Claude Code skill find the bugs What the original author did: ios-black-box works like a flight recorder for your iOS app: it records the screen, what the app says out loud and an event log of everything it did on a real iPhone, and a Claude Code skill reads the recording back and finds the bugs. Reported result (as claimed by the author): Open source; the builder's point is that the simulator guesses while the real phone shows what happened. Source: https://x.com/aydin_ayanzadeh/status/2111674935554777493 Tools they used: Claude Code Category: Dev & ops Please: 1. Ask me 3 short questions about my situation (accounts I have, budget, what I want the output to look like). 2. Propose a step-by-step plan using these tools or cheaper alternatives, with the time and cost of each step. Use official pricing pages; say "check pricing" if unsure. 3. Then do the setup with me one step at a time, waiting for my OK before anything that sends messages, posts publicly, spends money or deletes data. 4. Finish with a checklist to run it again next time.
You are helping me set up this AI use case end to end in Grok Bot. Goal: Record real iPhone sessions and let a Claude Code skill find the bugs What the original author did: ios-black-box works like a flight recorder for your iOS app: it records the screen, what the app says out loud and an event log of everything it did on a real iPhone, and a Claude Code skill reads the recording back and finds the bugs. Reported result (as claimed by the author): Open source; the builder's point is that the simulator guesses while the real phone shows what happened. Source: https://x.com/aydin_ayanzadeh/status/2111674935554777493 Tools they used: Claude Code Category: Dev & ops Please: 1. Ask me 3 short questions about my situation (accounts I have, budget, what I want the output to look like). 2. Propose a step-by-step plan using these tools or cheaper alternatives, with the time and cost of each step. Use official pricing pages; say "check pricing" if unsure. 3. Then do the setup with me one step at a time, waiting for my OK before anything that sends messages, posts publicly, spends money or deletes data. 4. Finish with a checklist to run it again next time.
The Daily 5
Get five new use cases tomorrow morning
Free. Real sources. Unsubscribe any time.
Guides for this use case
How to vibe code an app with Claude Code · How to build AI agents for SEO and ads · How to control coding agents from your phone · How to use Claude Code skills and plugins
Related use cases
Check on Claude Code and Codex sessions from your phone without SSH keys
For people running agents in tmux and checking on them with Termius and Tailscale, Tenax lets you sign in and see your sessions without SSH keys or a VPN. It pings you when an agent needs you or finishes.
Keep Claude Code sessions alive across your Mac, iPad and iPhone
A developer codes with Claude Code from a MacBook, iPad or iPhone against one Mac mini, but closing a window killed the session. He built Cockpit, which runs Claude in a daemon so conversations survive and follow him across devices.
Recover your Claude Code Desktop sessions after switching accounts
When you switch Claude accounts (for example personal to Team), the Code tab in Claude Desktop starts empty even though nothing was deleted. claude-migrate moves your Code sessions to the other account on the same Mac so they show in the sidebar again.
Compact Claude Code sessions when a task finishes
Built smart-compactor so Claude Code compacts context when a task is done instead of when the window is full.