claude-global-appbuilding-skills

Audit an existing application or guide a new build for AI governance — classifying every function as DETERMINISTIC, AI-ADVISORY, or AI-OPERATIONAL, identifying risk, and recommending hardening. Use when the user asks to "audit AI usage", "check governance", "review where AI is used", "assess AI risk", "governance review", "lock down AI", or when planning a new feature that involves AI and the user wants to understand where AI should vs. shouldn't be used. Also use when the user wants to evaluate whether a component should use AI or deterministic code.

Complete git workflow for a phase - branch, commit, merge, cleanup

Systematic root cause analysis for bugs and failures. Use when something is broken, failing, or behaving unexpectedly — not for quick typos or obvious one-line fixes. Trigger on "diagnose this", "root cause", "why is this broken", "this isn't working", "investigate this error", "what's causing this", "trace this issue". Also use when /execute-phase Layer 1 or Layer 2 fails twice on the same task.

Execute an approved implementation plan, with automated testing, UI verification, and acceptance-criteria review built into every task. Use this skill whenever the user is ready to start building — trigger on phrases like "execute phase X", "implement the plan", "let's start building", "start coding", "build this out", "time to code", "build phase X", "start phase X", "implement phase X", "let's build this", "start the build", "go ahead and build". Also trigger immediately after the user approves output from /plan-phase. Don't wait to be asked explicitly — if a plan exists and the user says something like "looks good, let's go", this is the right skill.

Guide a complex project ideation session through chunk identification, per-chunk discovery, and structured output files ready for /kickoff Pathway A

Deep market research and assumption mapping - second step in the ideation sequence

Explore a problem space and understand users - the first step in the ideation sequence

Generate solution hypotheses and prepare for kickoff - final step in the ideation sequence

Build a detailed execution plan for a specific numbered implementation phase — tasks, files to touch, acceptance criteria, and tests — structured for Claude Code to execute. Use this skill whenever the user wants to understand, plan, scope, or break down a phase before building it. Trigger on intent: planning what a phase contains, deciding if a phase is too big to tackle in one session, or preparing work to hand off to /execute-phase. Common phrasings include "plan/break down/scope/detail phase X", "what goes into phase X", "what needs to happen in phase X", "is phase X too big" — but also trigger on any casual variation that expresses wanting to understand or organize a phase's work before coding begins. Do NOT trigger when the user is ready to start coding (use /execute-phase) or is checking status mid-execution.

Walk through technology decisions for a new project collaboratively

Lightweight verification gate — prove a claim before making it. Use when finishing ad-hoc work, one-off fixes, or any time you're about to say "done" or "fixed" outside of /execute-phase. Trigger on "verify this works", "prove it", "check your work", "does this actually work". Also referenced inline by /diagnose (Phase 4) and /execute-phase (completion criteria) as an embedded methodology.

Initialize a new project by analyzing source documents and generating requirements.md and design.md

Comprehensive pre-deployment testing - dual E2E (Playwright/Cypress), MSW integration tests, conflict analysis, prioritized fix plan. Use for UAT, full test suite, pre-deployment testing.

Enforce codebase conventions and add data-testid attributes to interactive elements before E2E testing. Eliminates ~70% of E2E test failures from naming mismatches.