菜单
Write, fix, and run unit, integration, regression, browser, and end-to-end tests. Use when proving behavior, debugging failures, or improving coverage.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Explore options, trade-offs, and decision paths without implementing code. Use when the user asks to brainstorm, compare approaches, choose architecture, or save a durable decision note.
Plan and implement safe schema, migration, relationship, index, query, and backfill changes with rollback awareness. Use when data shape or persistence contracts change.
Manage deployment, Docker, CI/CD, environment, secrets, and infrastructure changes with rollback and verification. Use for operational config or release safety work.
Create, update, and audit durable project documentation. Use when code, setup, workflow, architecture, API behavior, or project structure must be reflected in docs.
Implement planned product or code changes phase by phase using repo patterns and verification. Use when the user asks to build, modify, or finish implementation work.
Coordinate multi-specialist work, handoffs, blockers, and durable progress artifacts. Use when tasks span roles, phases, or parallel workstreams.
| name | testing |
| description | Write, fix, and run unit, integration, regression, browser, and end-to-end tests. Use when proving behavior, debugging failures, or improving coverage. |
Use this skill for automated tests, coverage improvements, regression tests, and end-to-end or browser verification.
mcp-playwright.mcp-open-bridge for a second-pass test or diagnosis review.mcp-context7 for framework-specific testing APIsmcp-playwright for browser flows, snapshots, and UI checksmcp-github for CI failures or test-related PR contextmcp-open-bridge for flaky-test hypothesis review and alternative regression-test ideasInclude: