A developer has built an open-source mobile app that lets them remotely approve or deny blocked tool calls for AI agents running on their Mac, so work runs can continue even when they're away from their desk. The app uses Tailscale or home Wi-Fi to keep encryption keys local and connects via QR code pairing; Android is live and iOS is under review.
Summaries like this, in your inbox every morning.
Sign up free →What happened
A developer created an open-source mobile application that lets them approve or deny blocked tool calls for AI agents running on their Mac from their phone, using Tailscale or home Wi-Fi for connection. Android is live; iOS is in review.
Why it matters
AI coding agents often pause when they encounter gated tool calls (actions that need human permission), forcing developers to wait until they can return to their desk to restart runs. This app solves that bottleneck by enabling remote approval from a phone, keeping work flowing without moving encryption keys off the local machine.
What to watch
The developer is refining the first-run pairing experience (which requires a QR code setup) and welcomes feedback on the approval-routing architecture; iOS availability is pending review.
A software developer working with AI coding agents discovered a productivity bottleneck: whenever their agents encountered a gated tool call (an action that requires human approval before execution), the entire run would stall until they returned to their desk to manually approve or deny the action. To solve this, they built a mobile app available as open source that lets them manage these approvals from their phone while away from work. The app pairs with an agent gateway running on the developer's own Mac using either Tailscale (a private networking tool) or their home Wi-Fi network. A key design choice was to avoid cloud relays—instead, the connection is local and encryption keys never leave the Mac, addressing security and privacy concerns for developers running sensitive agent workloads. Setup uses QR code pairing to simplify the initial configuration. As of the submission, the Android version is live and available, while the iOS version is still under review. The developer has open-sourced the code and is actively seeking feedback, particularly on the approval-routing architecture and any friction users encounter during the first-run pairing process.
AI coding agents routinely encounter tool calls that require human approval before execution — for security, safety, or governance reasons. When a developer is away from their desk, these agents stall until the developer returns and manually approves the action, creating friction in remote or asynchronous workflows. This open-source tool addresses that specific pain point by moving the approval decision to a mobile device, keeping the agent run alive without the developer having to return to their machine. The design choice to use Tailscale or local Wi-Fi rather than a cloud relay reflects a privacy-conscious approach common in developer tools, where local control and key management matter to users running sensitive workloads. The developer's openness to feedback on pairing friction and the approval-routing architecture suggests this is still early—a pattern typical of open-source agent tools being built in the wild by developers solving their own immediate problems.
No comments yet. Be the first to share your thoughts!
Log in to join the discussionGet curated AI news from 200+ sources delivered daily to your inbox. Free to use.
Get Started FreeFree · takes 30 seconds · unsubscribe anytime
1 minute a day. The AI essentials.
200+ sources · Email / LINE / Slack
