Claude Code Agents

Expert in SwiftUI, iOS/macOS development, and glassmorphic UI design for the CubeSolver app

Single entrypoint that coordinates specialist agents for component delivery.

Architecture improvement agent based on deep modules. Use when: improving architecture, finding refactoring opportunities, module deepening, tightly-coupled modules, improve architecture, refactoring. Explores the codebase, identifies friction, and proposes module-deepening refactors as RFCs.

Specialized agent for deploying SecureBase multi-tenant AWS infrastructure using Terraform, with expertise in environment-specific configurations and multi-customer deployments.

Autonomous E2E evaluation orchestrator for the RALPH-style workflow loop. Executes the real workflow agents end to end, with live MCP-backed cost, Draw.io design, governance discovery, validation, and benchmark collection. Does NOT replace the production 01-Orchestrator.

Use after any significant implementation to update .context/ documentation. Keeps .context/ in sync with the actual current state of the codebase after changes.

Describe what this custom agent does and when to use it.

Documentation specialist for al-folio v1.x

Use when: endurecer scoring e machine learning, remover bootstrap sintético indevido, revisar champion challenger, governança de modelos, explainability, feedback loop, thresholds, inferência online e readiness de produção do ML.

Describe what this custom agent does and when to use it.

Desarrollador frontend para React 19, Tailwind CSS 4, páginas de dashboard y componentes UI glass-morphism. Usar para crear/modificar interfaces de usuario.

Use when asked to plan and coordinate multi-step test automation work, delegate tasks to specialized subagents, or when a request spans multiple approaches or requires exploration before writing. Orchestrates Explorer, Advanced Writer, POM Writer, Spaghetti Writer, and Reviewer agents.

Use when: reviewing code for bugs, analyzing code purpose and intent, suggesting improvements, checking code quality and security, ensuring documentation completeness

Improve and refine declarative agent instructions based on Microsoft best practices

Azure Container Apps specialist: deploys and scales containerized microservices, configures Dapr sidecars, manages revisions, and orchestrates serverless container workloads

Security-focused code review specialist with OWASP Top 10, Zero Trust, LLM security, and enterprise security standards

<!-- Inspired by awesome-copilot code review patterns --> --- description: "Comprehensive code review assistance focusing on quality, security, and PS2 theme consistency" tools: ['codebase', 'search', 'problems'] model: Claude Sonnet 4 ---

Plan or apply dependency and platform upgrades carefully in app-configs

Use when the user wants to DESIGN a new AI agent for a real-world domain or workflow — before the file is written. You research how practitioners actually do the work (web search, docs, OSS examples, in-repo patterns), distill a practical workflow brief, and translate it into a concrete agent spe...

Use when maintaining the AgToosa repository itself: generator shell flow, template wiring, platform config, AGTOOSA_VERSION, copy/update behavior, or bats tests. Keywords: agtoosa maintainer, agtoosa.sh, lib/, template/, tests/agtoosa.bats, release wiring.

Generates technical documentation, README updates, API docs, architecture diagrams, and walkthroughs for Vault. Never implements production code. Triggers: 'document', 'write docs', 'README', 'API docs', 'walkthrough', 'technical writing', 'diagrams'.

Implements backend source code for assigned wave tasks, invokes the UI agent for frontend tasks, runs verification, and opens a wave PR. Triggered by wave issues from the orchestrator workflow.