What problem does it solve? Multi-step development work—building features, fixing bugs, refactoring—often ships with regressions, band-aid fixes, and inconsistent quality because review, testing, and validation happen ad hoc. This Skill enforces a disciplined, iterative workflow that loops through analysis, building, review, testing, fixing, and simplification until every quality gate passes. ## Core Features & Use Cases - Iterative Quality Loop: Runs an 8-phase cycle (Analyse, Build, Review, Test, Fix, Simplify, Validate, Ship) that rejects band-aid fixes, breaking changes, and workarounds, iterating until all gates pass. - Multi-Agent Coordination: Delegates work to specialised solo subagents (code-explorer, code-architect, code-reviewer) and escalates to parallel agent teams (team-lead, team-implementer, team-reviewer, team-debugger) via /team-* slash commands for multi-stream or multi-dimensional work. - Automated Quality Scripts: Ships scripts for code review with band-aid detection, fix validation, code simplification, deployment readiness, and token budget monitoring. - Use Case: When asked to implement a new API endpoint, the Skill explores the codebase, designs the architecture, implements the change, runs parallel reviews across security/performance/correctness dimensions, executes tests, fixes root causes of any failures, and only ships once compatibility and quality thresholds are met. ## Quick Start Ask the AI to build, fix, or refactor a feature using the automate-dev workflow and let it iterate until all quality gates pass.
