Agent Skills by Brandon
Showing 36 vetted skills indexed across 1 GitHub repositories.
platform-awareness
Track Claude Code platform updates and generate change reports.
security
Identify security vulnerabilities in Apple platform applications across iOS, macOS, and watchOS.
architecture-patterns
Guides macOS Swift/SwiftUI architectural decisions using SOLID, DRY, and Clean principles.
ui-review-tahoe
Audit macOS Tahoe apps for Liquid Glass, HIG compliance, SwiftUI patterns, and accessibility.
macos-design-system
Standardize macOS SwiftUI UI work with centralized design tokens.
macos-best-practices
Define and enforce macOS development best practices for Swift projects.
macos-crash-recovery
Implement macOS crash recovery with NSApplication termination handling and Core Data autosave.
claude-tasks
Create, update, retrieve, and list Epics, Stories, and Tasks with persistent memory.
peekaboo
Automate macOS UI testing and visual QA via the Peekaboo MCP server.
swiftdata-architecture
Design SwiftData schemas, queries, repositories, and performance optimizations for iOS and macOS apps.
macos-dnd
Implements Transferable-based drag-and-drop for SwiftUI lists and cross-app file drops on macOS 13+.
macos-scenes
Structure macOS SwiftUI apps with multiple window scene types.
skill-creator
Create and bootstrap modular Claude skills with consistent frontmatter and templates.
github
Automate GitHub pull requests, reviews, and issue management via gh CLI.
swift-patterns
Enforce actor-based services, typed throws, and strict concurrency in Swift 6 macOS apps.
keychain
Store and retrieve credentials securely using macOS Keychain services.
xcode-build-patterns
Guide Xcode build settings, schemes, signing, and distribution for macOS apps.
core-data-patterns
Standardize Core Data persistence patterns for macOS apps with NSPersistentContainer.
xcode-mcp
Run Xcode build, test, and diagnostics workflows via the MCP bridge.
swiftui-patterns
Standardize macOS Tahoe SwiftUI project layout and architecture conventions.
story-context
Maintain cross-task storytelling context via a shared markdown file.
review-cycle
Encode the review cycle as a state machine with per-stage templates.
submit-for-review
Validate completed child tasks, tests, and builds before code review submission.
error-recovery
Generate structured recovery prompts for failed AI agent handoffs.