live-debugging-mcp

Diagnose iOS DAT app issues using live MCP debug server events and device state.

523|117|Updated Nov 21, 2025
One-click install
npx skills add https://github.com/facebook/meta-wearables-dat-ios --skill live-debugging-mcp
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: live-debugging-mcp
Source: https://github.com/facebook/meta-wearables-dat-ios/tree/main/plugins/mwdat-ios/skills/live-debugging-mcp
Command: npx skills add https://github.com/facebook/meta-wearables-dat-ios --skill live-debugging-mcp

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Debugging an iOS app built with the Meta Wearables Device Access Toolkit is hard when failures happen across the app, Meta AI companion app, and glasses boundary. This Skill gives an agent a structured, evidence-first workflow to diagnose registration, permission, session, and streaming failures using a local DAT Inspector MCP debug server instead of guessing from code alone.

Core Features & Use Cases

  • Live runtime diagnosis: Connect to a running app's debug server and inspect connection status, SDK state, and DAT readiness before changing any code.
  • Boundary and permission analysis: Inspect companion-app boundary diagnosis, device path, permissions, and device properties to separate iOS system issues from Meta AI app/device issues.
  • Event-driven evidence gathering: Wait for targeted events by category or source, collect errors and event digests, and export a redacted diagnostic bundle for handoff.
  • Use Case: A developer reports their app cannot stream from Meta AI glasses. The agent connects to the debug server, finds registration is blocked, and identifies a misconfigured Info.plist URL scheme and incomplete handleUrl(_:) callback as the cause.

Quick Start

Ask the agent to debug why your iOS DAT app cannot connect to the glasses using the local DAT Inspector MCP tools.

Frequently Asked Questions about live-debugging-mcp

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I debug an iOS Meta Wearables DAT app that cannot connect to glasses?

Connect the agent to the local DAT debug server, then check connection status, SDK state, and DAT readiness in that order. This separates MCP connectivity problems from registration, permission, session, and stream failures before any code changes.

How to diagnose DAT registration failures on iOS?

Verify the Info.plist DAT configuration, app URL scheme, MetaAppID, and that Wearables.shared.handleUrl(_:) processes the Meta AI callback. Live registration state from the debug server confirms whether the app reaches the registered state.

What is the difference between iOS permissions and DAT permissions?

iOS system permissions are granted by the operating system, while DAT permissions flow through the Meta AI app and device boundary. Use get_permissions and companion boundary diagnosis to check DAT permissions instead of assuming system permission status applies.

Why does my DAT stream fail after the session starts?

Inspect stream state, camera permission, and recent stream errors before changing frame processing code. Session success does not guarantee stream health, so gather targeted stream events and error evidence first.

Can this skill modify my app or device state during debugging?

No, the workflow is strictly read-only unless you explicitly request a code change. It never mutates Meta AI app, device, account, permission, registration, or app state during diagnosis.

What happens if the DAT Inspector MCP tools are not configured?

The agent asks you to enable the local DAT debug setup first. As a fallback, it relies on app logs plus DAT documentation search instead of live runtime evidence.