ccfg-core

Execute a beads task with structured workflow

Create a pull request with comprehensive pre-flight validation gates to ensure code quality and

Structured pull request review

Initialize project CLAUDE.md with language-specific conventions

Generate a comprehensive context bundle of the repository using repomix, making it easy to share the

Scan the workspace for secrets, API keys, hardcoded credentials, and common vulnerability patterns.

Use this agent when evaluating system architecture, validating design decisions, assessing scalability concerns, or reviewing technology stack choices. Examples: validating microservices design, reviewing database architecture, assessing API design patterns, evaluating caching strategies, analyzing service boundaries, reviewing infrastructure decisions, validating distributed system design, assessing architectural trade-offs.

Use this agent when developing server-side APIs, services, or business logic. Invoke for REST or GraphQL API implementation, database modeling, authentication systems, background jobs, or service integration. Examples: building payment processing service, implementing OAuth2 provider, creating webhook handlers, designing microservice APIs, or optimizing database queries.

Use this agent when configuring build systems, optimizing compilation performance, managing dependencies, setting up build caching, or troubleshooting build issues. Examples: configuring Webpack/Vite/Rollup, optimizing TypeScript compilation, setting up monorepo build orchestration, implementing build caching, managing package dependencies, optimizing Docker build layers, reducing bundle sizes, debugging build failures.

Use this agent when designing cloud architectures, implementing multi-cloud strategies, optimizing cloud costs, configuring IAM policies, or applying well-architected framework principles. Examples: designing AWS/Azure/GCP infrastructure, creating infrastructure-as-code templates, optimizing cloud spend, implementing zero-trust security models, designing disaster recovery strategies, configuring service meshes, implementing observability solutions, or migrating on-premises workloads to cloud platforms.

Use this agent when you need comprehensive code review covering quality, security, best practices, and conventions. Invoke for pull requests, security audits, architecture reviews, or before merging code. Examples: reviewing a new feature branch, auditing authentication logic, checking dependency updates, evaluating third-party code, or assessing technical debt.

Use this agent when diagnosing bugs, investigating production issues, analyzing stack traces, or performing root cause analysis. Invoke for unexpected behavior, performance degradation, test failures, or system errors. Examples: tracking down race conditions, analyzing memory leaks, investigating authentication failures, debugging API timeouts, or resolving integration issues.

Use this agent when investigating complex errors, analyzing log patterns, debugging distributed systems, detecting anomalies, or correlating errors across services. Examples: root cause analysis for production incidents, interpreting cryptic stack traces, identifying cascading failures in microservices, detecting error patterns from logs, analyzing performance degradations, investigating intermittent failures, or correlating errors across multiple systems to find underlying causes.

Use this agent when designing event-driven architectures, defining AsyncAPI specifications, planning event schemas, or architecting message-based systems. Invoke for topic/channel naming conventions, event envelope design, CloudEvents adoption, schema registry patterns, saga orchestration, dead letter queue strategies, consumer group design, or event versioning decisions.

Use this agent when building user interfaces, web components, or client-side applications. Invoke for React/Vue/Angular development, responsive design implementation, accessibility improvements, state management, or frontend performance optimization. Examples: creating reusable component library, implementing complex forms, building data visualization dashboards, optimizing bundle size, or ensuring WCAG compliance.

Use this agent when building complete features spanning database, backend, and frontend layers. Invoke for end-to-end feature implementation, full-stack architecture decisions, or integrating multiple system layers. Examples: building user authentication system, creating dashboard with data visualization, implementing real-time notifications, developing CRUD interfaces, or architecting new application modules.

Use this agent when establishing Git workflows, resolving merge conflicts, managing branches, reviewing Git history, or troubleshooting repository issues. Examples: setting up Git Flow or trunk-based development, resolving complex merge conflicts, interactive rebasing, managing release branches, cleaning up Git history, undoing commits, configuring branch protection, managing Git submodules, troubleshooting detached HEAD states.

Use this agent when designing GraphQL schemas, planning type systems, implementing Relay specifications, designing mutations, or architecting federated GraphQL services. Invoke for schema-first vs code-first decisions, nullability strategy, connection/pagination patterns, subscription design, query complexity analysis, persisted queries, or GraphQL security hardening.

Use this agent when designing gRPC services, defining proto3 schemas, planning streaming patterns, or architecting service-to-service communication. Invoke for protobuf message design, RPC method naming, backward-compatible schema evolution, gRPC error handling with rich details, deadline propagation, load balancing strategies, health checking protocol, or gRPC-Web gateway design.

Use this agent when analyzing performance bottlenecks, optimizing application speed, profiling code execution, reducing memory usage, or conducting load testing. Examples: identifying slow database queries, resolving N+1 query problems, optimizing API response times, reducing bundle sizes, fixing memory leaks, improving cache hit rates, analyzing profiling data, conducting load tests, optimizing rendering performance, reducing startup time.

Use this agent when decomposing epics into tasks, sequencing work, mapping dependencies, estimating scope, planning sprints, or organizing project delivery. Examples: breaking down large features into implementable stories, creating work breakdown structures, identifying critical path items, analyzing task dependencies, estimating effort and timeline, planning sprint capacity, identifying risks and mitigation strategies, or optimizing delivery sequences to maximize value and minimize blockers.

Use this agent when designing LLM prompts, optimizing prompt performance, creating evaluation frameworks for generative AI outputs, implementing prompt patterns, or improving prompt reliability. Examples: crafting effective system prompts for Claude agents, designing few-shot examples for consistent formatting, implementing chain-of-thought reasoning, creating evaluation rubrics for LLM outputs, A/B testing prompt variations, reducing hallucinations, or improving response quality through prompt engineering techniques.

Use this agent when you need test strategy design, quality assurance processes, test planning, risk-based testing approaches, exploratory testing guidance, defect triage, or quality metrics analysis. Examples: designing comprehensive test plans for new features, establishing testing frameworks, analyzing test coverage gaps, prioritizing test cases based on risk, investigating quality trends, setting up automated test pipelines, or evaluating testing ROI.

Use this agent when refactoring code to improve structure, readability, or maintainability while preserving behavior. Invoke for extracting functions, applying design patterns, removing duplication, simplifying complex logic, or modernizing legacy code. Examples: breaking up large functions, introducing dependency injection, converting callbacks to async/await, extracting reusable components, or applying SOLID principles.

Use this agent when eliciting requirements, writing user stories, defining acceptance criteria, managing scope, establishing traceability, or analyzing stakeholder needs. Examples: conducting stakeholder interviews to gather requirements, translating business needs into technical specifications, writing clear user stories with acceptance criteria, prioritizing features using MoSCoW or value vs effort analysis, creating requirements traceability matrices, identifying requirement conflicts or gaps, or facilitating requirement workshops with cross-functional teams.

Use this agent when designing REST APIs, modeling resources, defining HTTP semantics, creating OpenAPI specifications, or planning API versioning and evolution strategies. Invoke for resource URL architecture, pagination standards, caching strategies, CORS configuration, error response formats, bulk operations, long-running async patterns, idempotency key design, or API gateway considerations.

Use this agent when creating documentation, writing API references, producing user guides, documenting architecture decisions, or improving technical clarity. Examples: writing README files, creating API documentation, documenting configuration options, writing architecture decision records (ADRs), creating onboarding guides, documenting deployment procedures, writing changelog entries, creating inline code documentation, explaining complex technical concepts.

Use this agent when designing test strategies, setting up test frameworks, improving test coverage, implementing CI/CD testing pipelines, creating test fixtures, or debugging test failures. Examples: configuring Jest/Pytest/JUnit, writing unit tests, designing integration test suites, implementing e2e tests with Playwright/Cypress, setting up test databases, creating mock services, improving code coverage, parallelizing test execution, debugging flaky tests.

This skill should be used when testing web UIs, browser automation, end-to-end testing with Playwright or Puppeteer, Chrome DevTools debugging, or visual regression testing.

This skill should be used when persisting context between sessions, saving project state, loading previous session context, or managing longitudinal memory beyond beads issue tracking.

This skill should be used for ALL coding tasks, code reviews, planning, git operations, and development work. It defines mandatory workflow rules that must be followed in every session.

4 hooks across 4 events