From apple-dev
Use after executing a plan, or when the user says 'execution review', 'review implementation', 'check implementation completeness'. Compares plan vs code execution results, verifies implementation completeness and identifies deviations. Performs Apple platform Swift code scan (localization, concurrency, abstraction, error handling) and doc update checks.
npx claudepluginhub n0rvyn/indie-toolkit --plugin apple-devThis skill uses the workspace's default tool permissions.
Verify implementation matches plan. Use **after execution**.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Designs, implements, and audits WCAG 2.2 AA accessible UIs for Web (ARIA/HTML5), iOS (SwiftUI traits), and Android (Compose semantics). Audits code for compliance gaps.
Verify implementation matches plan. Use after execution.
Before invoking the implementation reviewer, search for known issues related to the modified modules:
search(query="<modified module names and component keywords>", source_type=["error", "lesson"], project_root="<cwd>")Use the dev-workflow:implementation-reviewer agent to perform the full plan-vs-code verification and design fidelity audit (if design doc exists).
Dispatch with:
docs/06-plans/ or user-specified)git diff --name-only or plan task file referencesThe agent returns a compact summary (verdict, gap counts, report file path) and writes the full report to .claude/reviews/. Wait for the agent to complete before proceeding to Step 2.
After the agent completes, perform these Swift-specific checks on all new or modified files:
String(localized:) — no hardcoded user-facing stringsString(localized:)await MainActor.run or marked @MainActor@MainActor@Model types not passed across actor boundaries!) on optionals from external sourcesVerify documentation was updated:
docs/04-implementation/file-structure.mddocs/07-changelog/YYYY-MM-DD.mddocs/03-decisions/Combine the implementation-reviewer agent's compact summary with Swift scan findings. Read the full report file for gap details when needed.
## Execution Review Summary
### Implementation Review
{agent compact summary — verdict, gap counts}
Full report: {report file path}
### Swift Code Scan
- Localization: {N} issues
- Concurrency: {N} issues
- Abstraction: {N} issues
- Error handling: {N} issues
### Doc Updates
- file-structure.md: ✅ updated / ❌ not updated
- changelog: ✅ updated / ❌ not updated
- ADR: ✅ created / ⚠️ not needed / ❌ missing
### Total
- Implementation gaps: {N}
- iOS issues: {N}
- Doc gaps: {N}