Notulensi

Use to implement a REST or GraphQL API endpoint from a specification or description, including request validation, error handling, and accompanying tests. Activate for building a new API route, resolver, or handler end-to-end. Do NOT use for API design/contract creation, documentation generation, or frontend API consumption.

Use to implement background job processing with a queue, worker, retry with exponential backoff, dead letter queue, and monitoring. Activate when the request involves offloading work from the request cycle (emails, reports, data processing). Do NOT use for real-time event streaming, webhook handling, or scheduled cron jobs.

Use to scaffold a new project from scratch — generating directory structure, configuration files, entry points, and a README skeleton. Activate when starting a new project, service, or package and needing a production-ready starting point. Do NOT use for adding features to existing projects, generating tests, or CI/CD configuration.

Use to design and implement a caching layer for data that is expensive to compute or fetch. Activate for cache-aside, write-through, or write-behind patterns, TTL strategy, and cache invalidation logic. Do NOT use for database query optimization, CDN configuration, or HTTP response caching headers.

Use to generate a CI/CD pipeline configuration for building, testing, and deploying an application. Activate for GitHub Actions, GitLab CI, CircleCI, or similar pipeline tools for any tech stack. Do NOT use for infrastructure provisioning, Dockerfile creation, or deployment strategy design.

Use to generate production-ready code for a new feature, function, class, or module from a description. Activate for writing new code from scratch with error handling, inline documentation, and a self-review checklist. Do NOT use for refactoring existing code, code review, or test generation.

Use to migrate existing code from one framework, language version, or technology stack to another. Activate for porting code with compatibility notes, breaking change identification, and maintaining test coverage through the migration. Do NOT use for refactoring within the same stack, dependency upgrades, or new feature development.

Use when a developer needs a thorough review of existing code before merging — covering bugs, security vulnerabilities, style violations, and improvement opportunities. Activate when the request involves reviewing a PR, function, or module for quality and correctness. Do NOT use for refactoring, generating new code, or writing tests.

Use to design and implement a configuration management system with environment separation, secrets handling, and startup validation. Activate when the request involves managing app config across dev/staging/production environments. Do NOT use for environment setup scripts, CI/CD configuration, or infrastructure secrets management.

Use when a developer has a slow or expensive database query and needs analysis, index recommendations, and a rewritten optimized version. Activate for query performance, EXPLAIN output interpretation, or index strategy for any relational or document database. Do NOT use for schema design from scratch, ORM configuration, or general database architecture.

Use to upgrade one or more dependencies to newer versions, including breaking change analysis, step-by-step migration instructions, and a rollback plan. Activate for library upgrades, framework version bumps, or security patch updates. Do NOT use for adding new dependencies, removing unused packages, or full stack migrations.

Use when a developer has a design problem and needs help identifying and implementing the most appropriate design pattern. Activate for structuring code around a recurring architectural challenge, applying OOP or functional patterns, or improving extensibility. Do NOT use for refactoring without a pattern goal, API design, or database schema design.

Use to containerize an application with an optimized Dockerfile and docker-compose configuration. Activate for creating Docker images, multi-stage builds, or local development container setups. Do NOT use for Kubernetes manifests, CI/CD pipeline configuration, or cloud container service setup.

Use to generate environment setup scripts, .env.example files, and README setup documentation for a project. Activate when onboarding a new developer or setting up a project for the first time on a new machine. Do NOT use for CI/CD pipeline configuration, Docker setup, or production infrastructure setup.

Implement a comprehensive error handling strategy with typed errors, propagation, logging, and user-facing messages. Do NOT use for debugging existing errors, writing tests, or logging-only instrumentation.

Implement a feature flag system for toggles, gradual rollout, and targeting. Do NOT use for configuration management, A/B test analytics, or infrastructure feature flags.

Generate a GraphQL schema for a domain, including types, queries, mutations, subscriptions, directives, and resolvers. Do NOT use for REST API design, OpenAPI spec generation, or GraphQL client code generation.

Generate Infrastructure as Code templates using Terraform, Pulumi, or AWS CDK. Do NOT use for Dockerfile creation, CI/CD pipelines, or application-level configuration.

Implements all methods of a defined interface, abstract class, or contract with full type safety, error handling, and documentation. Do NOT use for designing interfaces from scratch, API contract design, or generating tests only.

Adds structured logging to a module or service, including log levels, context fields, correlation IDs, and PII masking. Do NOT use for error handling design, metrics/tracing instrumentation, or log aggregation setup.

Use to configure a monorepo with workspace tooling, shared packages, build pipeline, and CI optimization. Activate for setting up Turborepo, Nx, Lerna, or similar monorepo tooling for a multi-package project. Do NOT use for single-package project setup, CI/CD pipeline scripting, or Docker configuration.

Use to generate an OpenAPI 3.x YAML specification for an API, either from existing code or from a description. Activate for creating API documentation, generating client SDKs, or defining an API contract. Do NOT use for GraphQL schema generation, API implementation, or Postman collection creation.

Use to implement pagination for an API endpoint or UI data list. Activate for cursor-based or offset pagination, total count, next/prev links, and performance considerations for large datasets. Do NOT use for infinite scroll UI implementation, search result ranking, or data filtering design.

Use when a developer has identified a performance issue in existing code and needs bottleneck analysis, optimized implementation, and benchmarks. Activate for slow code, high CPU/memory usage, or latency problems in a specific module. Do NOT use for database query optimization (use database-query-optimization skill), caching design, or infrastructure scaling.

Use to implement rate limiting middleware for API endpoints to prevent abuse and protect backend resources. Activate for configurable request limits, response headers, and bypass rules for trusted clients. Do NOT use for authentication, API gateway configuration, or DDoS protection at the infrastructure level.

Use to improve existing code quality — readability, maintainability, or performance — without changing its observable behavior. Activate for cleaning up code, reducing complexity, or applying better patterns to working code. Do NOT use for adding new features, fixing bugs, or generating code from scratch.

Use to plan and implement a migration from a REST API to GraphQL, including schema design, resolver mapping, deprecation strategy, and client migration guide. Activate for transitioning an existing REST API to GraphQL. Do NOT use for designing a new GraphQL API from scratch, REST API design, or API versioning.

Use to integrate a third-party SDK or library into their application with a proper abstraction layer, error handling, retry logic, and a mock for testing. Activate for adding a new external service integration. Do NOT use for building SDKs from scratch, API contract design, or webhook handling.

Performs a comprehensive production-readiness evaluation. Covers error handling, security, performance, observability, testing, and documentation with severity-rated findings. DO NOT use for quick style reviews, refactoring, or generating code.

Provides architectural guidelines, coding standards, folder structure, and best practices for 13 supported tech stacks. DO NOT use for generating actual code.

Implements a production-ready webhook receiver with signature verification, idempotency, retries, and dead letter queue. DO NOT use for outgoing webhooks or general endpoints.

Diagnoses API timeouts, identifies slow components via tracing, and provides retry/timeout strategies. DO NOT use for network infrastructure, DB query optimization, or general profiling.

Debug async/await errors, unhandled promise rejections, and event loop blocking. Do not use for synchronous logic bugs or race conditions in concurrent systems.

Debugs authentication token errors (JWT validation, OAuth token issues, expiry, signature, refresh flows). Do NOT use for CORS, session auth, or authorization logic.

Analyze build failures to identify root causes and provide fix instructions. Do not use for deployment failures or runtime errors.

Analyzes crashing, restarting, or OOMKilled Docker/Kubernetes containers. Fixes crash causes. Do NOT use for deployment failures, app bugs, or infrastructure.

Resolves browser CORS errors by identifying missing headers and applying server-side fixes. Do NOT use for authentication errors, API gateway config, or network firewalls.

Investigates incorrect, inconsistent, or missing data to identify the corruption source and create a remediation plan. Do NOT use for data migration planning, schema design, or general database optimization.

Interprets EXPLAIN/EXPLAIN ANALYZE output to diagnose slow queries and recommend indexes or rewrites. Do NOT use for schema design or ORM configuration.

Analyze application hangs, thread deadlocks, and circular wait conditions. Do not use for race conditions or infinite loops.