Loading...
Loading...
Found 9 Skills
Use when reviewing a plan before implementation begins. Not for autonomous plan analysis — use plan-review agent instead. Challenges scope, walks through architecture/quality/tests/performance interactively with mandatory user checkpoints and opinionated recommendations.
Performs comprehensive codebase audit checking architecture, tech debt, security vulnerabilities, test coverage, documentation, dependencies, and maintainability. Use when auditing a project, assessing codebase health, running security scans, checking for vulnerabilities, reviewing code quality, analyzing tech debt, or asked to audit/analyze the entire codebase.
Run technical quality checks across accessibility, performance, theming, responsive design, and anti-patterns. Generates a scored report with P0-P3 severity ratings and actionable plan. Use when the user wants an accessibility check, performance audit, or technical quality review.
Execute development tasks with skeleton-first approach and layered TDD. Includes optional adversarial verification (code quality + test completeness review). Use when users start working on a task (T-XX), need development guidance, or implement features/bugfixes. Triggers on keywords like "execute task", "start T-XX", "implement", "develop", "开发任务", "执行任务", "对抗式验证", "--review".
一个全面、结构化的特性开发工作流程，包含代码库探索、架构设计、质量审查和实现等专门阶段。在构建需要以下条件的新特性时使用：（1）在做出更改之前了解现有代码库，（2）通过系统性的提问澄清模糊的需求，（3）在实现之前设计周到的架构，（4）通过多维度的代码审查确保质量。非常适合涉及多个文件、需要架构决策、包含复杂集成或需求有些不明确的功能特性。
面向团队工作流的子代理开发技能（阶段3）。基于已批准的计划，按任务派发子代理实现，并执行“规格合规→代码质量”双阶段审查，必要时同步 Jira 子任务状态。
Use when user wants a pre-review cleanup pass on changed code by running parallel reviews for reuse, code quality, and efficiency, then applying worthwhile fixes without changing behavior.
Audit an ENTIRE codebase with multiple agents in parallel — map the repo, partition it into review slices, fan out one (or more, multi-lens) reviewer subagent per slice, reduce with a cross-cutting/architecture critic + a completeness check, then triage, fix, and report with an honest coverage ledger. Use when asked to 'audit the whole codebase', 'full security/quality review of the repo', 'review the entire project', 'do a deep/comprehensive code audit', 'scan everything for bugs or vulnerabilities', when onboarding/inheriting an unfamiliar repo, or for a periodic deep sweep. Whole-repo scoped and scales agent count to repo size — NOT a diff review (for changed lines use dual-agent-review).
Advise on Unity gameplay script quality