開發

Orchestrate lightweight parallel internet research (2-4 dimensions). Spawns light-research-researcher workers for each subtopic dimension, coordinates findings, synthesizes final reports. Use for standard research queries with 2-3 distinct angles. Examples - cloud gaming optimization, quantum computing overview, WebRTC performance analysis. Triggers - research, investigate, analyze with multiple aspects.

Modo Gerente de Projeto Ultra Realista - Estimativas conservadoras, crítica brutal, ROI calculado. TRIGGERS: 'brutal honesty', 'estimativa realista', 'vale a pena', 'ROI', 'crítica brutal'. SEMPRE ATIVO em project-context.

Apply SDD/TAC methodology for coding tasks. Auto-invoke when implementing features, refactoring code, fixing bugs affecting 3+ files, planning implementations, or when user mentions Scout-Plan-Build workflow. This skill guides systematic development through Phase 0 (Constraints), Scout (reconnaissance), Plan (design), and Build (execution) phases.

Manages state with Valtio using proxy-based reactivity, direct mutations, and automatic re-renders. Use when wanting mutable state syntax, fine-grained reactivity, or state management outside React components.

Evaluates whether a programming language dependency should be used by analyzing maintenance activity, security posture, community health, documentation quality, dependency footprint, production adoption, license compatibility, API stability, and funding sustainability. Use when users ask "should I use X or Y?", "are there better options for [feature]?", "what's a good library for [task]?", "how do we feel about [dependency]?", or when considering adding a new dependency, evaluating an existing dependency, or comparing/evaluating package alternatives.

Master ASP.NET Core fundamentals including C#, project structure, routing, middleware, and basic API development. Essential skills for all ASP.NET Core developers.

Coordinate with other AI agents running in tmux panes. Use this skill when you need to delegate tasks, request reviews, or collaborate with agents like Codex, Gemini, or other Claude instances.

Workflow for private repos + Colab training. Trigger when: (1) training on Colab with private repo, (2) separating dev from production, (3) avoiding untested code in production.

Use when designing API endpoints. Enforces layered architecture, separation of concerns, and type safety. Supports Express.js and similar frameworks.

Use when debugging NavigationStack issues (not responding, unexpected pops, crashes), build failures (SPM resolution, "No such module", hanging builds), or memory problems (retain cycles, leaks, deinit not called). Systematic diagnostic workflows for iOS/macOS.

Scaffolds new Buck2 projects with proper BUILD/PACKAGE files, SPDX headers, and depot shims. Use when creating new Rust binaries/libraries, Deno tools, or C++ projects in the monorepo. Ensures consistent structure and metadata from the start.

Development skill for CaseMark's Court Recording Transcriber - an AI-powered application for transcribing court recordings with speaker identification, synchronized playback, search, and legal document exports. Built with Next.js 16, PostgreSQL, Drizzle ORM, wavesurfer.js, and Case.dev APIs. Use this skill when: (1) Working on or extending the court-record-transcriber codebase, (2) Integrating with Case.dev transcription APIs, (3) Working with audio playback/waveforms, (4) Building transcript export features, or (5) Adding speaker identification logic.

Comprehensive PDF processing techniques for handling large files that exceed Claude Code's reading limits, including chunking strategies, text/table extraction, and OCR for scanned documents. Use when working with PDFs larger than 10-15MB or more than 30-50 pages.

Python 3.11+ development with type hints, async patterns, FastAPI, and pytest. Use for backend services, CLI tools, data processing, and API development.

Your approach to handling testing testing. Use this skill when working on files where testing testing comes into play.

MANDATORY protocol for executing implementation plans with independent tasks - dispatches fresh subagent for each task with code review between tasks, enabling fast iteration with quality gates. MANDATORY parallel execution for independent tasks.

Use this skill when creating next-generation autopoietic Claude skills with enhanced Git-brain integration, adaptive error healing, and emergent pattern recognition. Self-improved version that learned from v1's deployment, optimized for the MONAD ecosystem with φ-tier indexing.

Apply when making API calls to ChurchTools, using churchtoolsClient methods, handling API responses, or working with Tags API. Covers correct HTTP method names, parameter structure, and response handling patterns.

Use for any frontend UI implementation - orchestrates tailwind-ui, shadcn, and radix-ui in the correct order

Proactively suggests tests when code changes are detected. Activates when functions are created, logic is modified, security-sensitive code is added, or bug fixes are implemented. Provides non-intrusive test recommendations to help vibe coders ship with confidence.