t-tools

Plugin t-tools in timzaak/web-dev-skills. Commands read from the repo's manifest — 2 days ago.

Skills (33)

• t-decision

  Decide whether a feature should enter the t-tools workflow before tech research or PRD.

• t-design-check

  Evaluate technical design documents for implementability, completeness, and consistency with a quantitative 100-point score.

• t-design

  Generate technical design documents including API design, database schema, and implementation details for a feature.

• t-doc

  Generate human-readable project documentation — tutorials, getting-started guides, architecture overviews, API references, deployment or onboarding docs — by scanning the codebase for real facts. Output prefers an existing docs-web site over docs/tutorials/. Use when the user asks to write project d

• t-dream

  Organize and realign PRDs, user stories, design/task docs, implementation facts, and project structure so the target project keeps a clean current context instead of accumulating stale or misleading information. Use when asked to audit documentation drift, structure drift, traceability gaps, whether

• t-extension-demo-accept

  独立验收 Chrome 扩展真实加载后的 Playwright 用户故事演示并汇总报告；普通 Web 页面演示使用 t-web-demo-accept。

• t-extension-demo-run-all

  批量运行并修复 Chrome 扩展 Playwright 用户故事演示，维护独立批次报告和断点；普通 Web 批次使用 t-web-demo-run-all。

• t-extension-demo-run

  运行单个 Chrome 扩展 Playwright 用户故事演示，诊断并修复真实加载后的集成失败；普通 Web 页面演示使用 t-web-demo-run。

• t-figma-assets

  Prepare Figma bitmaps, SVGs, and videos for human-guided UI implementation; identify assets, validate their types, collect bitmap evidence, convert or optimize, and write a verified manifest. Does not implement UI.

• t-figma-impl

  Implement a Figma UI in an existing frontend file, block by block; leave pending assets empty, bind prepared assets, and hand over for manual acceptance. Does not prepare media assets.

• t-flutter-demo-accept

  Independently accept Android Patrol demos against user stories and execution evidence; use after Flutter demo development.

• t-flutter-demo-run-all

  Run all Android Patrol user-story demo files with resumable checkpoints.

• t-flutter-demo-run

  Run and repair one Android Patrol user-story demo file.

• t-how

  Interactive guide that teaches how to use the t-tools plugin. Explains the Decision -> PRD/Tech Research -> Design -> Super Run -> Demo -> Release workflow, maps the user's goal or question to the right /t-tools:t-* entry command, and shows preconditions and next steps. Use when the user types "/t-t

• t-html-show

  Generate or update an HTML Preview for a Markdown document.

• t-icon-prompt

  Turn a vague icon idea into a ready-to-paste generation prompt card through a short guided Q&A. Use when the user wants an app icon, logo mark, favicon, or marketing icon prompt for any image generator or an SVG-writing agent. Produces prompts only and never generates images. Not for in-app UI icon

• t-init

  Initialize a new Rust + React full-stack project; use --extension-demo to add Playwright integration-test infrastructure to an existing WXT extension; use --shopify to create a Shopify React Router app project (or retrofit an existing one) with the same demo/scripts infrastructure. Does not implemen

• t-prd-check

  Validate draft PRD, published PRD baseline, and user stories for quality and consistency.

• t-prd-publish

  After implementation and acceptance, summarize the draft PRD against delivered facts, then revise the existing formal PRD.

• t-prd

  Draft or update PRD and user stories, clarifying unresolved product choices. Use for requirements, not go/no-go decisions or technical design.

• t-push

  由 AI 基于 git diff 总结提交内容，使用脚本运行受影响区域 CI，通过后 git commit && git push。

• t-release

  Bump project version, create git tag, and push release commit to remote.

• t-review

  Review changed code for correctness bugs with parallel read-only finders and independent 3-state verification in small batches, isolating risky or complex candidates. Add specialist angles and a gap sweep for high-risk changes. Write an evidence-backed report; apply fixes only with --fix. Quality cl

• t-run

  Execute and resume item-based task plans generated by t-task for one delivery phase. Use after task planning; use t-super-run to combine planning and execution.

• t-simplify

  Review the changed code for reuse, simplification, efficiency, and altitude issues with 4 parallel read-only reviewers, then apply the fixes directly. The default scope is the upstream range plus uncommitted changes; a PR / branch / file target replaces it. Quality cleanup only, not bug hunting.

• t-skill

  Create or update a skill — by default inside the t-tools plugin repository; with --project inside the current target project's skill directory. Use when adding a new t-* command, turning a repeated workflow into a skill, or changing an existing skill's responsibility, structure, or trigger descripti

• t-super-run

  Plan and execute one specified delivery phase after technical design, with persisted recovery and independent read-only acceptance. Use to combine planning and execution; use t-task and t-run for separate plan review or item-based execution.

• t-task-check

  Validate task plan executability and consistency with a 100-point score and P0/P1/P2 fix list.

• t-task

  Convert technical design documents into executable phased task plans with ordered work breakdown.

• t-tech-research

  Assess technical feasibility for a feature. Scans codebase, checks dependencies, researches libraries, and writes a structured report under .ai/tech-research/.

• t-web-demo-accept

  Independently accept Web page demos against user stories and execution evidence, then summarize reports. Use after demo development; Chrome extension demos use t-extension-demo-accept.

• t-web-demo-run-all

  Discover all non-live Demo E2E tests and run them one file at a time in the main session, diagnosing and fixing failures inline, restarting the environment after backend code changes and rebuilding demo data between files, until every file is resolved or exhausts its retry budget. Supports an option

• t-web-demo-run

  Run a single demo E2E test file, diagnose failures, dispatch fixes to agents, and re-run until pass.

Subagents (40)

• backend-accept

  只读验收 Rust 后端的设计、运行证据和 API 契约；用于后端实现后的阶段收口。

• backend-consistency

  后端一致性检查专家（只读）。负责验证单个模块的 PRD 文档与后端代码实现的一致性。 触发场景： - 检查 PRD 与后端代码实现的一致性 - 验证 API 能力边界、数据模型、验证规则是否与 PRD 匹配 关键词：consistency check, PRD vs implementation, API boundary, data model validation - "检查 realm-user 模块 PRD 与实现一致性" - "验证用户管理模块的 PRD 与代码匹配度" - "一致性检查 billing 模块" - Read - Grep - Glob - Bash

• backend-design

  Rust 后端技术设计专家。负责生成分端后端设计文档，拥有 API 契约的单一设计权，产出可直接进入 /t-super-run 的后端设计。

• backend-dev

  Rust 后端开发专家。负责后端生产代码、通过价值门槛的最小必要测试和来自 backend/test 集中 runner 的生产代码缺陷修复。

• backend-test

  后端场景测试编写专家。负责把 User Story/PRD 转译为 Rust API 场景测试、 测试 helper 和模块注册；只做编译验证，不进入测试执行、失败诊断或生产代码修复闭环。 单元测试由 backend-dev 负责；测试执行与修复编排由 backend/test 集中 runner 负责。 在任务规划中，负责把 backend/test 拆为测试编写（authoring）与集中执行（runner）两个职责。 - Read - Edit - Write - Grep - Glob - Bash - AskUserQuestion - mcp__context7__resolve-

• context-curator

  只读审计 PRD、用户故事、设计文档中的重复、过期、冲突与过程化信息，输出权威文档建议和整理计划。

• extension-accept

  只读验收扩展权限、上下文和运行证据；故事演示交给 extension-demo-accept。

• extension-demo-accept

  只读验收真实加载的 Chrome 扩展故事演示；Web 页面演示交给 web-demo-accept。

• extension-demo-dev

  编写和修复 Chrome 扩展真实加载后的 Playwright 用户故事演示；普通 Web 页面演示交给 web-demo-dev，扩展实现交给 extension-dev。

• extension-demo-diagnose

  只读诊断 Chrome 扩展 Playwright 演示失败，区分测试资产、扩展实现、后端和环境问题，输出结构化报告。

• extension-design

  为 Chrome MV3 / WXT 扩展变更生成 extension.md；普通 Web 页面交给 frontend-design。

• extension-dev

  基于 WXT + React 实现或修复 Chrome MV3 扩展代码、权限和跨上下文功能；不负责独立测试资产或只读验收。

• extension-test

  编写和修复 Chrome 扩展的 Vitest 业务逻辑与组件测试；Playwright 扩展演示交给 extension-demo-dev，生产代码缺陷交回 extension-dev。

• figma-impl

  Figma 整页 UI 实现者。按块 baseline、context 和 assets manifest 实现页面并运行栈验证。

• flutter-accept

  只读验收 Flutter 架构、测试与平台证据；故事演示交给 flutter-demo-accept。

• flutter-demo-accept

  只读验收 Android Patrol 故事演示与设备运行证据。

• flutter-demo-dev

  基于用户故事编写和修复 Android Patrol 演示测试。

• flutter-demo-diagnose

  只读诊断 Patrol 演示失败并输出结构化证据。

• flutter-design

  为 Flutter 功能或技术变更生成 flutter.md，明确体验、状态、平台行为和验证交接；不负责实现、测试执行或验收。

• flutter-dev

  按目标项目锁定版本实现和修复 Flutter View、Riverpod 状态、数据层与平台能力。

• flutter-test

  编写和验证 Flutter 单元、Widget、integration_test 与 Patrol 测试。

• frontend-accept

  只读验收前端行为证据、类型质量和 API 一致性；完整故事演示交给 web-demo-accept。

• frontend-design

  React Web 技术设计专家；负责页面、组件及 TanStack Query/Zustand 状态分工的分端设计，只消费后端 API 契约不重新定义。Chrome 扩展设计交给 extension-design。

• frontend-dev

  前端开发专家。负责 React 管理后台功能实现与前端缺陷修复。 触发场景： - 编写或修改 frontend 代码 - 实现页面、表单、表格、共享组件 - 集成 API、路由、缓存、前端交互 - 修复前端构建、类型、交互或 Demo 暴露的问题 关键词：frontend, react, component, page, form, table, tanstack router, react query, tailwind - Read - Edit - Write - Grep - Glob - Bash - AskUserQuestion - WebSearch - mcp__contex

• frontend-test

  前端测试专家。负责 React 管理后台的 Vitest 组件测试和集成测试编写， 使用 @testing-library/react 与 MSW 做隔离测试。 触发场景： - 编写或修改 frontend 测试 - 为组件、hooks、schema、局部交互补 Vitest - 补充 MSW handlers、fixtures、测试工具 - 修复前端单测或集成测试失败 关键词：frontend test, vitest, testing-library, msw, component test, integration test - Read - Edit - Write - Grep -

• html-show

  将 Markdown 文档转为 HTML Preview，支持交互原型、流程图、状态图与能力矩阵。

• miniapp-accept

  只读验收小程序构建、技术线与微信业务运行证据；用于实现后的阶段收口。

• miniapp-dev

  微信小程序开发专家，基于 Taro + React 实现页面、组件、主题接入与缺陷修复。

• miniapp-test

  需要独立测试或专项验证资产时，编写和验证 miniapp 测试、模板门禁及相关构建配置。

• review-finder

  只读从调度方指定的正确性角度审查代码变更，或在高风险首轮验证后查漏；返回带触发场景的候选缺陷。

• review-verifier

  在独立于 finder 的上下文只读验证一批 correctness review 候选，对每项给出三态判定与证据；未完成检查的项显式返回缺口。

• shopify-accept

  只读验收 Shopify App/扩展交付面及运行证据；故事演示交给 web-demo-accept。

• shopify-design

  Shopify App 技术设计专家；负责 React Router 应用本体、app proxy、theme app extension 与 customer account UI extension 的分端设计。Shopify 项目无独立 backend 端时由本角色产出 app proxy / internal API 契约。

• shopify-dev

  Shopify App 开发专家。负责 React Router 应用本体（loader/action、Prisma、admin 嵌入页、app proxy）、 theme app extension 与 customer account UI extension 的实现与缺陷修复。 触发场景： - 编写或修改 Shopify 应用本体代码 - 实现 admin 嵌入页、app proxy 路由、数据模型迁移 - 实现店面 app block、顾客账户扩展页面 - 修复 Shopify 构建或运行问题 关键词：shopify, react router, polaris, app bridge

• shopify-test

  Shopify App 测试专家。负责应用本体纯逻辑（验签函数、loader/action 内可独立提取的业务规则） 与扩展辅助模块的 Vitest 测试编写。 触发场景： - 为 app proxy 验签、业务规则、状态转换、数据访问边界等纯逻辑补测试 - 修复 Shopify Vitest 测试失败 - 补充测试 fixture/helper 关键词：shopify test, vitest, hmac verify, pure logic - Read - Edit - Write - Grep - Glob - Bash - AskUserQuestion - mcp__context

• simplify-reviewer

  只读审查本次代码变更的一个清理角度（复用/简化/效率/抽象层级），输出带具体代价的结构化 findings；具体角度由调度方注入。

• structure-review

  只读评估项目目录结构、模块边界、测试布局与 .ai 产物组织是否合理。

• web-demo-accept

  只读验收 Web 页面故事演示及整文件执行证据；扩展演示交给 extension-demo-accept。

• web-demo-dev

  Demo 测试开发专家。基于用户故事和设计文档生成或修复 Playwright E2E 演示测试。 在需要编写 demo/e2e 测试、从用户故事生成测试代码，或修复 Demo 测试失败时使用。 关键词：demo test, playwright e2e, user story test, selector calibration, demo/e2e - Read - Write - Edit - Glob - Grep - Bash - AskUserQuestion

• web-demo-diagnose

  Demo 测试诊断专家。只做失败诊断、问题分类和诊断报告输出，不修改业务代码。 触发场景： - Demo 测试失败后需要定位根因 - 需要判断问题属于测试代码、前端、后端、权限、数据还是环境 - 需要生成结构化诊断报告供后续修复 agent 使用 关键词：demo diagnose, test failure analysis, playwright error, selector failure, api failure, timeout - Read - Glob - Grep - Bash - Write