delivery-loop

Build inclusive, WCAG-compliant UIs. Use for a11y, screen readers, ARIA, keyboard nav, contrast, or accessibility audits.

Analyze a codebase and generate Project_Architecture_Blueprint.md — detected stacks and architecture patterns, diagrams, layers, cross-cutting concerns, extension guidance. Use to document, map, or blueprint a project's architecture or keep it consistent.

Continuous backend delivery loop — cycles subagents through test → diagnose → fix → review → secure → re-test until the suite, quality, and security gates are clean. Use to develop, fix, harden, or finish a backend feature/service/endpoint.

Design backend systems — REST/GraphQL/gRPC APIs, microservices, database schemas, auth, caching, message queues, scalability. Use when designing backend services, APIs, data models, distributed systems, or auth flows.

Shared backend guide for Langfuse's Next.js/tRPC/BullMQ/TypeScript monorepo — tRPC routers, public REST endpoints, BullMQ processors, services, Prisma/ClickHouse access, Zod validation, and backend tests across web/worker/shared.

Turn backend test ambiguity into one practical test packet — API/service/repo/auth coverage, fixture & seed/reset strategy, mock-vs-container choices, contract checks, and flaky-suite stabilization across local and CI.

Create API handoff documentation for frontend developers. Use when backend work is complete and needs to be documented for frontend integration, or user says 'create handoff', 'document API', 'frontend handoff', or 'API documentation'.

Guide for building modern, accessible, and composable UI components. Use when building new components, implementing accessibility, creating composable APIs, setting up design tokens, publishing to npm/registry, or writing component documentation.

BullMQ Redis-backed job queues in Node.js/TypeScript — workers, delayed/repeatable jobs, flows, rate limiting, retries. Use for background jobs, job scheduling, or reliable async processing.

Create beautiful visual art in .png and .pdf documents using design philosophy. Use when the user asks to create a poster, piece of art, design, or other static visual piece.

The bar for any code you write, edit, or review. Read these rules whenever you are writing or reading any code

React composition patterns that scale — compound components, render props, context providers, lifting state. Use when refactoring boolean-prop-heavy components or designing flexible, reusable component APIs (incl. React 19)

Write and apply Conventional Commits — type(scope): subject, imperative mood, with body and BREAKING CHANGE footer when needed. Use when committing staged work or asked to write, review, or standardize a commit message.

Core Web Vitals reference for measuring, diagnosing, and improving LCP, INP, and CLS page performance against Google's thresholds

Create professional SVG logos, brand identities, icons, wordmarks, and visual marks — multiple concepts, layouts, and color variations. Use when asked to design a logo, brand identity, icon, or visual mark.

Organize CSS using BEM, SMACSS, and CSS-in-JS patterns. Use when building scalable, maintainable styling systems with proper naming conventions.

Creates, updates, and fixes Cypress tests (E2E and component) — use to write, add, update, or fix failing/flaky tests, even if the user only says "tests". For explaining or reviewing without edits, prefer cypress-explain.

Search and extract Cypress information from official documentation (docs.cypress.io, cypress.io); prefer LLM markdown under /llm/* and refuse unverified API or behavior claims.

Explains Cypress tests (E2E and component) and how Cypress works — use to understand, review, or critique tests without editing code, even if the user only says "tests". For creating or fixing tests, prefer cypress-author.

Write a structured retrospective of a delivery-loop run (frontend-delivery-loop, backend-delivery-loop, or fullstack-delivery-loop) that judges what went well, what went badly, where time/roundtrips/tool calls were wasted and in which skill or agent, how the agents integrated, and the session id + trace file paths — then save it as a Markdown report for skill-evolution-factory to act on. Use this the moment a delivery loop finishes (or stalls), whenever the user says "retro", "post-mortem", "retrospectiva", "feedback loop", "evaluate the loop", "how did the loop go", "what should we improve in the loop/agents", or asks to grade/critique a frontend/backend/fullstack delivery-loop run. Run it even if the user only gestures at it — a finished loop is the trigger.

Disciplined diagnosis loop for hard bugs and performance regressions — reproduce → hypothesise → instrument → fix → regression-test. Use when asked to diagnose/debug, something is broken or failing, or a perf regression is reported.

Develop, configure, and debug Fastify (Node.js/TypeScript) servers and REST APIs — routes, plugins, JSON Schema validation, hooks, auth, serialization, performance, deployment. Triggers: Fastify, server.ts, app.ts, fastify.config.

Create or update Figma Code Connect template files (.figma.ts/.figma.js) that map Figma components to code. Use for Code Connect, Figma component mapping, design-to-code.

Continuous frontend delivery loop — cycles subagents through test → diagnose → fix → review → re-test until E2E and review are clean. Use to develop, fix, harden, or finish a frontend feature/epic verified by tests.

Front-end system-design reference — box model, rendering/reflow, DOM & Web APIs, virtualisation, state design, network transports, web performance.

Continuous fullstack delivery loop — cycles a combined frontend+backend team through test → diagnose → fix → review → secure → re-test, building both layers together in cross-side vertical-slice waves until every E2E/suite, code-quality, visual-fidelity, and security gate is clean. Use to develop, fix, harden, or finish a feature that spans the API and the UI in one coordinated loop, instead of running the frontend and backend loops back-to-back.

Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".

Hunting playbook for race-condition / TOCTOU vulnerabilities — concurrent-request abuse, HTTP/2 single-packet attacks, double-redeem / double-spend, MFA-OTP & email-verify races, rate-limit bypass. Use when hunting races or TOCTOU bugs.

Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states

Create software diagrams in Mermaid — class, sequence, flowchart, ERD, C4, state, gitgraph, gantt. Use to "diagram", "visualize", "map out", or "show the flow" of architecture, database schemas, code structure, or user/app flows.

NestJS best practices — 40 rules across modules, DI, error handling, security, performance, testing, database, API design, microservices, and deployment. Use when writing, reviewing, or refactoring NestJS code.

Next.js App Router principles. Server Components, data fetching, routing patterns.

Next.js performance optimization for Core Web Vitals, image/font optimization, caching, streaming, bundle size, and Server Components. Use when fixing LCP/INP/CLS, implementing next/image or next/font, configuring unstable_cache, converting Client to Server Components, streaming with Suspense, or reducing bundle size. Covers Next.js 16 + React 19 patterns.

Build production-ready Node.js backend services with Express/Fastify — middleware, error handling, auth, database integration, and API design. Use when creating Node.js servers, REST APIs, GraphQL backends, or microservices.

Review code and architectures against the OWASP Top 10:2025. Use to audit for vulnerabilities, guide remediation, or write secure code. Triggers: "security review", "OWASP audit", "check for vulnerabilities", "is this code secure".

Performance optimization specialist — profiling, caching, latency, N+1, connection pools, p99 tail latency, async, load-testing. Use when performance, slow queries, profiling, caching, optimization, or database latency are mentioned.

Write, debug, and maintain Playwright tests — flaky tests, Page Object Model, locators/waits, fixtures, network mocking, auth, CI/CD. Covers E2E, component, API, visual, a11y, and extension testing.

Review SQL, schemas, and migrations for PostgreSQL best practices — JSONB, arrays, ENUMs/domains, PL/pgSQL, RLS, privileges. Use when reviewing any PostgreSQL DDL, query, function, or migration.

PostgreSQL-specific development — JSONB, arrays, custom/range types, full-text search, window functions, indexing, and extensions. Use when writing, tuning, or modeling anything on PostgreSQL.

Full API lifecycle via the Postman MCP Server — sync OpenAPI specs to collections, generate typed clients, run API tests, create mocks, publish docs, OWASP audits. Use for Postman, collections, specs, mocks, or SDKs.