full-stack-developmentlisted

Orchestrate production full-stack feature delivery across product contracts, frontend, backend, data, IAM, API, security, testing, observability, CI, and runtime evidence. Use when implementation spans multiple application layers or an accepted feature must be carried end to end. Routes only the smallest relevant specialist set, preserves planning and release approval gates, and keeps UI, API, authorization, and persistence contracts aligned. Not for a single settled specialist task, product discovery alone, or generic architecture.
ongkipro/dotfiles · ★ 0 · API & Backend · score 53

Install: claude install-skill ongkipro/dotfiles

# Full-stack Development Own the delivery thread, not specialist implementation. Start from repository truth and accepted behavior, route each concern to its canonical owner, keep cross-layer contracts synchronized, and stop only when the requested outcome has executable evidence or an explicit external blocker. Read [End-to-end workflow](references/end-to-end-workflow.md) for the runnable phase checklist, contract record, stack combinations, and evidence packet. Read [source-ledger.md](references/source-ledger.md) only when a current platform or standard claim is needed. ## Trigger and scope Use this skill when work crosses two or more of UI, server/runtime, data, IAM, API, telemetry, or delivery; when an accepted feature needs end-to-end execution; or when a cross-layer bug must be traced to its source. Route directly to one specialist when the task is already bounded to one domain. This skill does not author a generic architecture, repeat framework recipes, create project templates, or replace specialist review. It owns selection, sequencing, contract alignment, approval boundaries, and evidence gates. ## Inspect before selecting skills 1. Read repository instructions, accepted product/spec artifacts, and relevant status records. Then inspect the package manager and scripts, installed versions, routes, entry points, schemas and migrations, API contracts, IAM, tests, telemetry, deployment configuration, and existing design tokens. 2. Trace the affected flow