Token导航 LogoToken导航

Agent Skill · 研究检索

implement-feature(实现功能)

implement-feature 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

implement-feature 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为操作浏览器;许可证为 MIT,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 51,212。

总安装 588周安装 24GitHub Stars 86下载 188核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:implement-feature(实现功能)
来源仓库:https://github.com/marcelmichau/fake-survey-generator
仓库路径:skills/implement-feature
安装命令:
npx skills add https://github.com/marcelmichau/fake-survey-generator --skill implement-feature
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/marcelmichau/fake-survey-generator --skill implement-feature

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

implement-feature 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词快速定位候选结果。适用于研究类任务,如查找技术方案、筛选参考资料或匹配开发需求。通过 npx skills add 命令安装,需结合来源 README 核验具体用法。

使用前建议确认权限范围和是否会触发联网或文件操作。适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

来源仓库内路径:skills/implement-feature

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/marcelmichau/fake-survey-generator --skill imp…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要操作浏览器,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

implement-feature 怎么安装?

在终端执行 npx skills add https://github.com/marcelmichau/fake-survey-generator --skill implement-fea…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

implement-feature 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

implement-feature 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

implement-feature 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

implement-feature 用的人多吗?

公开数据是累计安装 588 次,近一周安装 24 次,下载 188 次,GitHub 86 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

implement-feature 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
implement-feature

实现功能 · 本页

58886操作浏览器MIT通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Implement Feature Skill

Use this skill to implement a new feature end-to-end with complete validation from code changes through interactive E2E testing.

What This Skill Does

This is an orchestrator skill that sequences the following steps:

  1. Analyze Requirements - Understand the feature request and identify affected components
  2. Implement Changes - Generate code changes across backend/frontend as needed
  3. Build Backend - Compile.NET backend projects
  4. Build Frontend - Compile TypeScript/React frontend
  5. Run Tests - Execute unit and integration tests to validate logic
  6. Start Aspire Runtime - Spin up the complete application environment
  7. Validate E2E - Execute feature-specific happy-path validation in running UI
  8. Report Results - Document validation outcomes (success/failure)

This orchestrator halts at the first failure and reports details to the user for remediation.

When to Use

Use this skill when:

  • Implementing a new feature from requirements to completion
  • Need end-to-end validation that feature works correctly
  • Want a guided, checkpoint-based development workflow
  • Need to track progress through multiple validation stages

How the Agent Should Use This Skill

Phase 1: Requirements Analysis

Input: Feature requirement or story from the user

Agent Task:

  1. Analyze the feature request to understand:

- What user-facing behavior is expected? - What components need changes (Frontend, API, Application logic)? - What data/models are involved? - What happy-path user journey should be validated?

  1. Document your understanding:

- Affected backend services (Api, Application, Worker) - Frontend components to modify (React components, pages) - Database changes needed (new entities, migrations) - API endpoints to create or modify - Happy-path validation flow (e.g., "user clicks button → form displays → data saved → list updated")

  1. Checkpoint: Confirm understanding with user before proceeding to implementation

Example Analysis:

Feature: Change survey creation button to purple
Analysis:
- Frontend Component: `src/client/ui/src/components/SurveyList.tsx` (button styling)
- Backend Changes: None required
- Happy-Path: Navigate to survey list → verify button is purple

Phase 2: Implementation

Agent Task:

  1. Implement backend changes (if any):

- Create/modify API endpoints in src/server/FakeSurveyGenerator.Api/ - Create/modify business logic in src/server/FakeSurveyGenerator.Application/ - Create/modify domain models as needed - Add/update database migrations if schema changes

  1. Implement frontend changes (if any):

- Modify React components in src/client/ui/src/ - Update types/interfaces for new API contracts - Update UI styling/layout as needed - Handle new validation/error scenarios

  1. Add/Update Tests:

- Unit tests in src/server/FakeSurveyGenerator.Application.Tests/ - Integration tests in src/server/FakeSurveyGenerator.Api.Tests.Integration/ - Aim for coverage of happy-path and edge cases

  1. Checkpoint: Review changes are complete before proceeding to validation

Implementation Checklist:

  • Backend logic implemented (if needed)
  • Frontend UI updated (if needed)
  • Tests written for new logic
  • No syntax errors in code
  • Dependencies updated if needed (NuGet, npm)

Phase 3: Build Validation

Agent Task: Invoke Build Backend Skill

Invoke: build-backend skill
Expected: Exit code 0, no compilation errors
On Failure: Report compilation errors to user, stop

Agent Task: Invoke Build Frontend Skill

Invoke: build-frontend skill
Expected: Exit code 0, no TypeScript errors
On Failure: Report build errors to user, stop

Phase 4: Test Validation

Agent Task: Invoke Test Backend Units Skill

Invoke: test-backend-units skill
Expected: All tests pass, exit code 0
On Failure: Report failed test names and assertions, stop

Checkpoint: All tests must pass before proceeding to E2E

Phase 5: Runtime Validation

Agent Task: Invoke Validate Aspire Runtime Skill

Invoke: validate-aspire-runtime skill
Expected: All resources healthy, UI endpoint discovered
Steps:
1. Run `aspire run` command in a terminal
2. Use Aspire MCP list_resources tool to monitor health
3. Wait for all resources (sql, cache, api, worker, ui) to be healthy
4. Extract UI endpoint URL (typically https://localhost:3000)
5. Report endpoint to next skill

On Failure: Report which resource failed to start, stop

Important: Leave Aspire running after this step; it's needed for E2E validation

Phase 6: E2E Validation

Agent Task: Invoke Validate E2E Skill

Invoke: validate-e2e skill
Context Provided to E2E Skill:
- UI endpoint URL from Aspire runtime
- Feature requirements/happy-path flow from Phase 1

E2E Skill Execution:
1. Use Playwright MCP to navigate UI
2. Execute feature-specific validation steps
3. Verify expected state changes and behavior
4. Report pass/fail with evidence

Expected: All validations pass
On Failure: Report specific validation failures, stop

Phase 7: Results & Cleanup

Success Path ✅:

  1. Report to user: "Feature implementation validated successfully!"
  2. Provide summary:

- Builds: Passed ✅ - Tests: N tests passed ✅ - E2E Validation: Feature working as expected ✅

  1. User can stop Aspire (Ctrl+C) when ready

Failure Path ❌:

  1. Report to user: "Validation failed at [specific stage]"
  2. Provide details:

- Stage: [Build/Test/Runtime/E2E] - Error: [specific error details] - Recommendation: [fix guidance if applicable]

  1. User must fix issues and re-run validation from failing stage

Checkpoint & Halt Strategy

The orchestrator uses the following halt behavior:

StageHalt ConditionAction
Requirements AnalysisN/AConfirm understanding
ImplementationUser indicates incompleteHalt, ask for completion
Backend BuildExit code ≠ 0Halt, report compilation errors
Frontend BuildExit code ≠ 0Halt, report build errors
Backend TestsAny test failsHalt, report test failures
Aspire RuntimeResources don't reach healthyHalt, report startup error
E2E ValidationAssertions failHalt, report validation failures
SuccessAll stages passComplete ✅

Typical Feature Implementation Timeline

  1. Analysis - 5-10 minutes
  2. Implementation - 15-60 minutes (depends on feature complexity)
  3. Builds - 1-2 minutes
  4. Tests - 1-3 minutes
  5. Aspire Startup - 2-5 minutes (first run slower)
  6. E2E Validation - 2-10 minutes
  7. Total - 30-90 minutes end-to-end

Common Feature Types & Implementation Patterns

Type: New API Endpoint

  • Backend Changes: Create DTO, handler, controller action
  • Frontend Changes: Create API client call, UI components
  • Tests: Unit test handler logic, integration test endpoint

Type: Frontend UI Component Change

  • Backend Changes: Usually none
  • Frontend Changes: Modify React component, styles
  • Tests: Snapshot tests, interaction tests in acceptance tests

Type: Database Feature (e.g., new entity)

  • Backend Changes: Create entity, update DbContext, add migration
  • Frontend Changes: Add UI for new entity (if user-facing)
  • Tests: Repository/handler tests with real database (Testcontainers)

Type: Cross-Cutting (spans frontend + backend + database)

  • All layers: Implement complete flow end-to-end
  • Tests: Full integration test from API to database
  • E2E: Happy-path user journey through complete workflow

Reference: Available Skills

Each phase invokes a specific skill:

PhaseSkill NameStandalone Skill Location
Build Backendbuild-backend.github/skills/build-backend/
Build Frontendbuild-frontend.github/skills/build-frontend/
Run Teststest-backend-units.github/skills/test-backend-units/
Start Runtimevalidate-aspire-runtime.github/skills/validate-aspire-runtime/
E2E Validationvalidate-e2e.github/skills/validate-e2e/

For more details on each skill, see their individual SKILL.md files.

Foundation Knowledge

Reference: fake-survey-gen-foundation skill for detailed project architecture, service topology, and patterns.

Example: Complete Feature Implementation

Scenario: "Add ability to search surveys by title"

Phase 1: Analysis

Feature: Search surveys by title
Changes Needed:
- Backend: Add search parameter to GetSurveys API endpoint
- Frontend: Add search input box to survey list page
- Happy-Path: List surveys → type search term → verify filtered results

Phase 2-4: Implement & Build

- Add `searchTerm` parameter to Application service method
- Add filtering logic (e.g., LINQ .Where(s => s.Title.Contains(searchTerm)))
- Add React input field with onChange handler
- Call API with search parameter
- Write unit tests for search logic
- Write integration test for endpoint with search
- Builds pass ✅
- Tests pass ✅

Phase 5: Runtime

- Start aspire run
- Resources reach healthy ✅
- UI endpoint: https://localhost:3000

Phase 6: E2E Validation

- Navigate to https://localhost:3000
- See survey list with multiple surveys
- Type "test" in search box
- Verify only surveys with "test" in title remain
- Clear search, verify full list returns
- Validation passes ✅

Result: Feature implemented, tested, and validated end-to-end ✅

平台分布

Claude Code

26.16%
按下载量换算49

Gemini CLI

23.73%
按下载量换算45

Antigravity

18.71%
按下载量换算35

windsurf

13.2%
按下载量换算25

trae

8.39%
按下载量换算16

Codex

3.66%
按下载量换算7

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。

继续浏览同类 Skills