shihwesleys-harness

Design consultation: understands your product, researches competitors, proposes a complete design system (aesthetic, typography, color, layout, spacing, motion), and generates font+color preview pages. Creates DESIGN.md as your project's design source of truth. Use when asked to "design my app", "create a design system", "pick fonts and colors", "design consultation", or "make it look good". For existing sites, use /plan-design-review to infer the system instead.

Post-ship documentation update. Reads all project docs, cross-references the diff, updates README/ARCHITECTURE/CONTRIBUTING/CLAUDE.md to match what shipped, polishes CHANGELOG voice, cleans up TODOS, and optionally bumps VERSION. Use when asked to "update docs", "sync documentation", "document what shipped", "update the readme after shipping", or "post-release docs".

Eng manager-mode plan review. Lock in the execution plan — architecture, data flow, diagrams, edge cases, test coverage, performance. Walks through issues interactively with opinionated recommendations. Use when asked to "review my plan", "eng review", "check my architecture", "review before I build", or "sanity check this plan".

Pre-landing PR review. Analyzes diff against the base branch for SQL safety, LLM trust boundary violations, conditional side effects, and other structural issues. Use when asked to "review my code", "review this PR", "check before merge", "pre-landing review", or "look at my diff".

Ship workflow: detect + merge base branch, run tests, review diff, bump VERSION, update CHANGELOG, commit, push, create PR. Use when asked to "ship it", "ship this branch", "merge and deploy", "create a PR", "land this", or "push to production".

Fast headless browser for QA testing and site dogfooding. Navigate any URL, interact with elements, verify page state, diff before/after actions, take annotated screenshots, check responsive layouts, test forms and uploads, handle dialogs, and assert element states. ~100ms per command. Use when you need to test a feature, verify a deployment, dogfood a user flow, or file a bug with evidence.

App Store Connect automation for iOS/visionOS: REST API, JWT auth, TestFlight, metadata management, build upload, submission flow. Use when automating any part of the App Store release pipeline.

App Store Connect CLI (asc) patterns, commands, and CI setup. Use when automating App Store submissions, TestFlight distribution, metadata, screenshots, or signing via the asc CLI tool.

Google Photorealistic 3D Tiles + RealityKit/visionOS integration. Use when working on globe apps, 3D city rendering, tile streaming, or ECEF coordinate transforms on Apple platforms.

iOS/visionOS commercial app release process — App Store prep, legal/compliance, monetization, marketing, technical readiness, launch, post-launch. Use when preparing an app for public release.

iOS/visionOS release pipeline patterns for solo/small team: branching, versioning, CI/CD, submission automation, phased rollout, hotfixes, changelogs, post-release monitoring. Use when shipping an iOS or visionOS app update.

Automated plan execution pipeline. Use when running /orchestrate on an /interactive-planning output. Handles plan ingestion, review, skill matching, tech research, user gating, and phased execution with worktree isolation, testing, code review, and incremental commits.

Trigger self-learning synthesis on demand. Analyzes recent session learnings, clusters patterns, and updates MEMORY.md with verified insights. Use when user says /self-learn, wants to consolidate learnings, or needs to manually trigger the synthesis that normally runs on session end.

Use when designing state machines, verifying state transition logic, auditing existing state enums for impossible states or deadlocks, or when interactive-planning specs describe stateful behavior. Also use when user says "verify this state machine", "check for deadlocks", "tla", "formal verification", or "prove this design".

Unity cross-platform XR patterns for visionOS + Meta Quest from one codebase. Use when building Unity apps targeting both Apple Vision Pro and Quest — covers package setup, XR rig pattern, input abstraction, Cesium tile budgets, version pinning, and known gotchas.

visionOS patterns, APIs, and spatial computing conventions. Use when building apps for Apple Vision Pro — provides scene types, RealityKit ECS, ARKit providers, gesture handling, and visionOS 26 updates.

visionOS game engine patterns and tradeoffs. Use when choosing between Unity PolySpatial, Unreal Engine, Godot, or native Metal for visionOS development — provides engine capabilities, rendering paths, simulator support, and integration patterns.

Create and edit JSON Canvas files (.canvas) with nodes, edges, groups, and connections. Use when working with .canvas files, creating visual canvases, mind maps, flowcharts, or when the user mentions Canvas files in Obsidian.

Create and edit Obsidian Bases (.base files) with views, filters, formulas, and summaries. Use when working with .base files, creating database-like views of notes, or when the user mentions Bases, table views, card views, filters, or formulas in Obsidian.

Interact with Obsidian vaults using the Obsidian CLI to read, create, search, and manage notes, tasks, properties, and more. Also supports plugin and theme development with commands to reload plugins, run JavaScript, capture errors, take screenshots, and inspect the DOM. Use when the user asks to interact with their Obsidian vault, manage notes, search vault content, perform vault operations from the command line, or develop and debug Obsidian plugins and themes.

Structured simulator debugging for iOS/visionOS apps. Use when user says "debug this on simulator", "the app is doing X wrong", "run the app and check", "sim-debug", "why is the UI showing X", "something is broken in the simulator", "figure out why X happens", or describes visual/behavioral bugs in a running app. Also use when resuming with "resume debugging" or "pick up where we left off". Do NOT use for build errors without runtime behavior, unit test failures without simulator, or general code questions.