AI-AGENT-SKILLS

Build UIs that work for all users including keyboard navigation, screen readers, and WCAG 2.2

Design multi-agent systems with robust tool interfaces, state management, and failure handling

Build ML systems with disciplined training, evaluation, deployment, and safety practices

Design APIs that are stable, ergonomic, and evolvable

Design systems at the right scale with explicit trade-off documentation

Design services that are reliable, observable, secure, and maintainable

Design CI/CD pipelines with fast feedback, quality gates, and reliable deployments

Conduct staff-engineer-level code reviews across correctness, security, performance, readability, and architecture

Remove accidental complexity without changing behavior — apply Chesterton's Fence

Manage what goes into the AI agent context window for maximum quality and minimum waste

Build data pipelines with contracts, lineage, quality gates, and observability

Schema design, migrations, indexing, and query patterns for maintainable and performant databases

Systematic root-cause analysis — stop guessing, start isolating

Deploy with rollback plans, staged rollouts, and feature flags

Write documentation that stays accurate and decisions that stay recorded

Challenge assumptions and hidden dependencies before they become bugs

Build UIs that are fast, accessible, maintainable, and tested

Branching strategy, commit discipline, PR hygiene, and semantic versioning

Sharpen a vague idea into a buildable, scoped proposal

Structured incident response: detect, communicate, mitigate, resolve, and prevent recurrence

Ship working vertical slices instead of big-bang implementations

Manage infrastructure with version-controlled, reviewed, and tested IaC

Systematically evaluate LLM outputs, prevent regressions, and measure quality

Design service boundaries, contracts, and failure modes for distributed systems

Build iOS/Android apps that are performant, offline-capable, accessible, and battery-efficient

Design logging, metrics, tracing, and alerting that make systems debuggable in production

Profile before optimizing; optimize with evidence, not intuition

Decompose work into atomic, independently verifiable tasks

Write prompts that are reliable, testable, versioned, and safe

Design, evaluate, and operate retrieval-augmented generation systems

Extract, validate, and document requirements from raw input

Identify and mitigate technical and delivery risk before it becomes an incident

Apply security controls, threat modeling, and hardening to code and infrastructure

Ground implementation in authoritative sources — specs, RFC, official docs — not assumptions

Write a specification before writing any implementation code

Write tests before implementation to drive design and prove correctness

Discover and compose skills from this framework to guide your work