long-running-harness

Decompose a large system specification into cook-it-ready epic beads via DDD bounded contexts, with optional post-epic improvement loop configuration

Multi-agent review with lesson-calibrated reviewers, runtime verification, and severity classification

Comprehensive playbook for building world-class websites, landing pages, web applications, and dashboards. Covers the full build sequence from brand identity and information architecture through layout, typography, color, scroll animations, micro-interactions, hover effects, loading states, empty states, error handling, accessibility, SEO, and conversion optimization. Use this skill whenever building a new website or web app from scratch, redesigning pages (pricing, landing, portfolio), adding polish or animations, fixing generic-looking UI, improving visual hierarchy, or when the user says anything looks "janky", "generic", "amateur", or "needs more polish". Also use when adding hover states, scroll effects, page transitions, or doing a design review pass.

Reflect on the cycle and capture high-quality lessons for future sessions

Decompose work into small testable tasks with clear dependencies

Develop precise specifications through Socratic dialogue, EARS notation, and Mermaid diagrams

Team-based TDD execution with adaptive complexity and agent delegation

Audit and set up a codebase for agentic AI development using the 15-principle manifesto

Hands-on QA testing of running applications via browser automation. Use whenever the user asks to test a running app, perform exploratory testing, verify UI/UX behavior, check accessibility, test forms and inputs, inspect network requests, do a smoke test, or systematically try to break things. Also use when review needs runtime verification beyond automated tests -- visual checks, boundary inputs, interaction flows, and design review against a live application. Triggers on phrases like "test this app", "QA this", "check the UI", "try to break it", "smoke test", "does this look right", "test the form", "check accessibility", "poke at it", or any request to manually verify a running web application or API.

Multi-phase test suite optimization with adversarial review

Reviews code for architectural compliance and design integrity

Checks implementation for drift from established constraints

Reviews test quality, assertions, and edge case coverage

Deep research producing structured survey documents for informed decision-making