Software skills for AI agents

Adversarial implementation review methodology that validates code completeness against requirements with fresh objectivity. Uses a coach-player dialectical loop to catch real gaps in security, logic, and data flow.

Design, document, and test APIs following RESTful principles, consistent conventions, and comprehensive testing strategies. Use this skill when the user asks about designing an API, writing API documentation, testing endpoints, handling API versioning, error response design, or wants guidance on building APIs that are intuitive, reliable, and maintainable.

Design software systems with sound architecture — choosing patterns, defining boundaries, making trade-offs, and planning for scalability. Use this skill when the user asks about system design, architecture decisions, design patterns, scaling strategy, service boundaries, or technical planning. Also trigger for "how should I structure this," "design a system for," "monolith vs microservices," "what pattern should I use," "scalability plan," or any request involving high-level technical design, component decomposition, or architectural trade-off analysis.

Perform deep, actionable code reviews covering bugs, security vulnerabilities, performance issues, and maintainability concerns. Use this skill when the user asks to review code, check a pull request, audit code quality, get feedback on an implementation, or wants a second pair of eyes on their work. Also trigger for "review this," "what's wrong with this code," "can you check my PR," "code quality check," or any request for constructive criticism of source code. Works with any language or framework.

Optimize database performance through indexing strategies, query optimization, schema design, and capacity planning. Use this skill when the user asks about slow queries, database indexing, query optimization, schema design for performance, database scaling, or wants guidance on diagnosing and fixing database bottlenecks.

Design database schemas, optimize queries, plan migrations, and develop indexing strategies for relational and NoSQL databases. Use this skill when the user asks about database design, schema modeling, query performance, SQL optimization, migrations, indexing, or data modeling. Also trigger for "slow query," "design a schema," "normalize this," "add an index," "migration strategy," "N+1 problem," or any task involving database architecture, query tuning, or data storage decisions. Covers PostgreSQL, MySQL, SQLite, MongoDB, Redis, and general database principles.

Methodical debugging — reproduce, isolate, root-cause, and fix bugs using systematic techniques rather than guesswork. Use this skill when the user reports a bug, encounters an error, gets unexpected behavior, sees a stack trace, or needs help troubleshooting. Also trigger for "why is this broken," "this doesn't work," "I'm getting an error," "help me fix this," "unexpected output," or any situation where code isn't behaving as intended. Works with any language, framework, or runtime environment.

Design and implement CI/CD pipelines, Docker configurations, infrastructure-as-code, and deployment strategies. Use this skill when the user asks about CI/CD, Docker, containers, Kubernetes, deployment, infrastructure, DevOps practices, or build pipelines. Also trigger for "set up a pipeline," "Dockerize this," "deploy this app," "GitHub Actions workflow," "automate deployment," "infrastructure as code," or any task involving build automation, containerization, orchestration, or production infrastructure. Covers GitHub Actions, GitLab CI, Docker, Kubernetes, Terraform, and general DevOps principles.

Write clear, comprehensive technical documentation — API references, READMEs, architecture decision records, onboarding guides, and inline code documentation. Use this skill when the user asks to write documentation, create a README, document an API, write an ADR, explain a system, or create developer guides. Also trigger for "document this," "write a README," "API docs," "how do I explain this codebase," "onboarding guide," or any task involving technical writing for a developer audience. Adapts to any language, framework, or documentation tooling.

Convert mistakes into executable rules using a structured error-to-rule system. Covers error categorization, rule extraction, pre-decision scanning, and maintenance for continuous improvement.

Use this skill when designing or restructuring a TypeScript application that spans a React frontend and a Node backend. Activate when users mention typescript react node backend architecture, fullstack architecture, monorepo layout, API layer design, shared types, end-to-end type safety, tRPC vs REST vs GraphQL, Next.js app structure, service layering, or "where should this code live" questions in a TS codebase.

Comprehensive NestJS best practices covering architecture, dependency injection, security, performance, testing, database patterns, API design, microservices, and deployment across 40 rules prioritized by impact.

Profile, identify bottlenecks, and optimize application performance across the full stack — frontend, backend, database, and infrastructure. Use this skill when the user asks about performance optimization, slow responses, profiling, load testing, caching strategy, or resource efficiency. Also trigger for "make this faster," "this is slow," "optimize performance," "reduce load time," "memory leak," "CPU usage is high," or any task involving measuring and improving application speed, throughput, or resource consumption. Language and framework agnostic.

Systematically refactor code using safe transformation patterns — improving structure, readability, and maintainability without changing behavior. Use this skill when the user asks to refactor, restructure, clean up, simplify, decompose, or reorganize code. Also trigger for "this code is a mess," "make this more maintainable," "reduce complexity," "extract this into," "break this up," or any request to improve code structure while preserving existing functionality. Works with any language or framework.

Perform comprehensive security audits covering OWASP top 10 vulnerabilities, dependency CVEs, secrets exposure, and threat modeling. Use this skill when the user asks for a security review, vulnerability assessment, penetration test review, threat model, or security hardening guidance. Also trigger for "is this secure," "check for vulnerabilities," "security best practices," "audit this for security," or any concern about application security, data protection, or attack surface reduction. Works with any language, framework, or infrastructure.

Generate secure, optimized SQL queries with proper parameterization, input validation, injection prevention, performance optimization, and cross-database syntax awareness for PostgreSQL, MySQL, SQLite, and SQL Server.

Auto-generate comprehensive test suites — unit tests, integration tests, and end-to-end tests — for any language or framework. Use this skill when the user asks to write tests, add test coverage, create a test suite, generate specs, or anything involving automated testing. Also trigger when the user says "add tests for this," "test this function," "increase coverage," "write specs," or mentions TDD, BDD, property-based testing, or test-driven development. Detects the project's language, test framework, and conventions automatically from the codebase.

Apply version control best practices for branching strategies, commit hygiene, collaboration workflows, and repository management. Use this skill when the user asks about git workflows, branching strategies, commit message conventions, merge vs rebase, monorepo management, or wants guidance on organizing their version control for solo or team development.