Observability Agent Skill

An agent skill that helps AI coding agents like Claude Code, Codex, Cursor, and Gemini write correct Swift observability code on Apple platforms - unified logging that respects privacy and persistence, signpost instrumentation that shows up in Instruments, MetricKit metrics and diagnostics from production, crash reports you can actually symbolicate, and telemetry pipelines that do not lie to you.

It uses the [Agent Skills](https://agentskills.io/home) format, so it works smoothly with Claude Code, Codex, Gemini, Cursor, and more.

What It Covers

• Unified logging - Logger (iOS 14+) and legacy os_log, subsystem/category design, the eight level methods and which levels persist to disk, privacy redaction (.public, .private, .private(mask: .hash)), formatting/alignment, lazy interpolation and why String-taking wrappers destroy it, multi-destination logging facades
• Log retrieval - Console.app filtering and saved searches, the log CLI (stream/show/collect/config with predicate recipes), logging configuration profiles, sysdiagnose capture, the Xcode 15+ structured console, and scripted three-tier capture for agent-assisted debugging (simulator stream-to-file, devicectl console launch with an env-gated print mirror, timestamped sudo log collect device archives that include app extensions)
• OSLogStore - reading your own logs at runtime