Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问许可证需确认审计未展示

build构建

Agent Skill

用于辅助数据库表结构、查询语句、迁移脚本和数据维护任务。它适合让 Agent 分析 schema、编写 SQL、排查查询问题、整理索引或生成迁移建议。使用时需要明确数据库类型、连接环境和目标表,区分只读分析与写入变更;涉及删除、更新、迁移和批量导入时,应优先 dry-run、备份或事务保护,避免误操作。

总安装

1,058

周安装

45

GitHub Stars

6,093

下载量

371
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

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:build(构建)
来源仓库:https://github.com/redisearch/redisearch
仓库路径:skills/build
安装命令:
npx skills add https://github.com/redisearch/redisearch --skill build
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。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/redisearch/redisearch --skill build

简介

用于 Redisearch 模块编译和构建流程。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。

  • 适合本地开发和测试环境搭建。build 属于研究检索类 Skill,可作为该场景下的辅助能力补充。
  • 通过 npx 命令从 GitHub 仓库安装使用。
  • 需确认权限范围和维护状态,注意可能触发命令执行。
  • 建议结合原始 README 核验具体用法和功能边界。

SKILL.md

Feature development pipeline - research, plan, track, and implement major features.

Instructions

This command manages a 4-phase feature development workflow for building major features. Parse $ARGUMENTS to determine which subcommand to run.

Arguments provided: $ARGUMENTS

Argument Parsing

Parse the first word of $ARGUMENTS to determine the subcommand:

  • research [name] → Run the Research phase
  • implementation [name] → Run the Implementation phase
  • progress [name] → Run the Progress phase
  • phase [n] [name] → Run Phase n of the implementation
  • status [name] → Show current status and suggest next step
  • (empty or unrecognized) → Show usage help

If the feature name is not provided in arguments, you MUST use AskUserQuestion to prompt for it.


Subcommand: Help (empty args)

If no arguments provided, display this help:

/build - Feature Development Pipeline

Subcommands:
  /build research [name]        Deep research on a feature idea
  /build implementation [name]  Create phased implementation plan
  /build progress [name]        Set up progress tracking
  /build phase [n] [name]       Execute implementation phase n
  /build status [name]          Show status and next steps

Example workflow:
  /build research chat-interface
  /build implementation chat-interface
  /build progress chat-interface
  /build phase 1 chat-interface

Then use AskUserQuestion to ask what they'd like to do:

  • question: "What would you like to do?"
  • header: "Action"
  • multiSelect: false
  • options:

- label: "Start new feature research" description: "Begin deep research on a new feature idea" - label: "Continue existing feature" description: "Work on a feature already in progress" - label: "Check status" description: "See what step to do next for a feature"


Subcommand: research

Step 1: Get Feature Name

If feature name not in arguments, use AskUserQuestion:

  • question: "What's a short identifier for this feature? (lowercase, hyphens ok - e.g., 'chat-interface', 'user-auth', 'data-export'). Use 'Other' to type it."
  • header: "Feature name"
  • multiSelect: false
  • options:

- label: "I'll type the name" description: "Enter a short, kebab-case identifier for the feature"

Step 2: Check for Existing Research

Check if docs/{name}/RESEARCH.md already exists.

If it exists, use AskUserQuestion:

  • question: "A RESEARCH.md already exists for this feature. What would you like to do?"
  • header: "Existing doc"
  • multiSelect: false
  • options:

- label: "Overwrite" description: "Replace existing research with fresh exploration" - label: "Append" description: "Add new research below existing content" - label: "Skip" description: "Keep existing research, suggest next step"

If "Skip" selected, suggest running /build implementation {name} and exit.

Step 3: Gather Feature Context

Use AskUserQuestion to understand the feature:

  • question: "Describe the feature you want to build. What problem does it solve? What should it do? (Use 'Other' to describe)"
  • header: "Description"
  • multiSelect: false
  • options:

- label: "I'll describe it" description: "Provide a detailed description of the feature"

Step 4: Research Scope

Use AskUserQuestion:

  • question: "What aspects should the research focus on?"
  • header: "Focus areas"
  • multiSelect: true
  • options:

- label: "Technical implementation" description: "APIs, libraries, architecture patterns" - label: "UI/UX design" description: "Interface design, user flows, interactions" - label: "Data requirements" description: "What data to store, schemas, privacy" - label: "Platform capabilities" description: "OS APIs, system integrations, permissions"

Step 5: Conduct Deep Research

Now conduct DEEP research on the feature:

  1. Codebase exploration: Understand existing patterns, similar features, relevant code
  2. Web search: Research best practices, similar implementations, relevant APIs
  3. Technical deep-dive: Explore specific technologies, libraries, frameworks
  4. Use AskUserQuestion FREQUENTLY: Validate assumptions, clarify requirements, get input on decisions

Research should cover:

  • Problem definition and user needs
  • Technical approaches and trade-offs
  • Required data models and storage
  • UI/UX considerations
  • Integration points with existing code
  • Potential challenges and risks
  • Recommended approach with rationale

Step 6: Write Research Document

Create the directory if needed: docs/{name}/

Write findings to docs/{name}/RESEARCH.md with this structure:

# {Feature Name} Research

## Overview
[Brief description of the feature and its purpose]

## Problem Statement
[What problem this solves, why it matters]

## User Stories / Use Cases
[Concrete examples of how users will use this]

## Technical Research

### Approach Options
[Different ways to implement this, with pros/cons]

### Recommended Approach
[The approach you recommend and why]

### Required Technologies
[APIs, libraries, frameworks needed]

### Data Requirements
[What data needs to be stored/tracked]

## UI/UX Considerations
[Interface design thoughts, user flows]

## Integration Points
[How this connects to existing code/features]

## Risks and Challenges
[Potential issues and mitigation strategies]

## Open Questions
[Things that still need to be decided]

## References
[Links to relevant documentation, examples, articles]

Step 7: Next Step

After writing the research doc, inform the user:

"Research complete! Document saved to docs/{name}/RESEARCH.md

Next step: Run /build implementation {name} to create a phased implementation plan."


Subcommand: implementation

Step 1: Get Feature Name

If feature name not in arguments, use AskUserQuestion to prompt for it (same as research phase).

Step 2: Verify Research Exists

Check if docs/{name}/RESEARCH.md exists.

If it does NOT exist:

  • Inform user: "No research document found at docs/{name}/RESEARCH.md"
  • Suggest: "Run /build research {name} first to create the research document."
  • Exit

Step 3: Check for Existing Implementation Doc

Check if docs/{name}/IMPLEMENTATION.md already exists.

If it exists, use AskUserQuestion:

  • question: "An IMPLEMENTATION.md already exists. What would you like to do?"
  • header: "Existing doc"
  • multiSelect: false
  • options:

- label: "Overwrite" description: "Create a fresh implementation plan" - label: "Append" description: "Add new phases below existing content" - label: "Skip" description: "Keep existing plan, suggest next step"

If "Skip" selected, suggest running /build progress {name} and exit.

Step 4: Read Research Document

Read docs/{name}/RESEARCH.md to understand:

  • The recommended approach
  • Technical requirements
  • Data models needed
  • UI/UX design
  • Integration points

Step 5: Design Implementation Phases

Break the research into practical implementation phases. Each phase should:

  • Be independently valuable (deliver something usable)
  • Be small enough to complete in a focused session
  • Build on previous phases
  • Have clear success criteria

Use AskUserQuestion to validate phase breakdown:

  • question: "How granular should the implementation phases be?"
  • header: "Phase size"
  • multiSelect: false
  • options:

- label: "Small phases (1-2 hours)" description: "Many focused phases, easier to track progress" - label: "Medium phases (half day)" description: "Balanced approach, moderate number of phases" - label: "Large phases (full day)" description: "Fewer phases, each delivering significant functionality"

Step 6: Conduct Phase Research

For each phase you're planning, do targeted research:

  • Web search for implementation specifics
  • Review relevant code in the codebase
  • Identify dependencies between phases

Use AskUserQuestion for any uncertainties about phase ordering or scope.

Step 7: Write Implementation Document

Write to docs/{name}/IMPLEMENTATION.md with this structure:

# {Feature Name} Implementation Plan

## Overview
[Brief recap of what we're building and the approach from research]

## Prerequisites
[What needs to be in place before starting]

## Phase Summary
[Quick overview of all phases]

---

## Phase 1: [Phase Title]

### Objective
[What this phase accomplishes]

### Rationale
[Why this phase comes first, what it enables]

### Tasks
- [ ] Task 1
- [ ] Task 2
- [ ] Task 3

### Success Criteria
[How to verify this phase is complete]

### Files Likely Affected
[List of files that will probably need changes]

---

## Phase 2: [Phase Title]

[Same structure as Phase 1]

---

[Continue for all phases]

---

## Post-Implementation
- [ ] Documentation updates
- [ ] Testing strategy
- [ ] Performance validation

## Notes
[Any additional context or decisions made during planning]

Step 8: Next Step

After writing the implementation doc, inform the user:

"Implementation plan complete! Document saved to docs/{name}/IMPLEMENTATION.md

Next step: Run /build progress {name} to set up progress tracking."


Subcommand: progress

Step 1: Get Feature Name

If feature name not in arguments, use AskUserQuestion to prompt for it.

Step 2: Verify Implementation Doc Exists

Check if docs/{name}/IMPLEMENTATION.md exists.

If it does NOT exist:

  • Inform user: "No implementation document found at docs/{name}/IMPLEMENTATION.md"
  • Suggest: "Run /build implementation {name} first."
  • Exit

Step 3: Check for Existing Progress Doc

Check if docs/{name}/PROGRESS.md already exists.

If it exists, use AskUserQuestion:

  • question: "A PROGRESS.md already exists. What would you like to do?"
  • header: "Existing doc"
  • multiSelect: false
  • options:

- label: "Overwrite" description: "Start fresh progress tracking" - label: "Keep existing" description: "Keep current progress, suggest next step"

If "Keep existing" selected, read the progress doc and suggest the next incomplete phase.

Step 4: Read Implementation Document

Read docs/{name}/IMPLEMENTATION.md to extract:

  • All phase titles
  • Tasks within each phase
  • Success criteria

Step 5: Create Progress Document

Write to docs/{name}/PROGRESS.md with this structure:

# {Feature Name} Progress

## Status: Phase 1 - Not Started

## Quick Reference
- Research: `docs/{name}/RESEARCH.md`
- Implementation: `docs/{name}/IMPLEMENTATION.md`

---

## Phase Progress

### Phase 1: [Title from Implementation]
**Status:** Not Started

#### Tasks Completed
- (none yet)

#### Decisions Made
- (none yet)

#### Blockers
- (none)

---

### Phase 2: [Title]
**Status:** Not Started

[Same structure]

---

[Continue for all phases]

---

## Session Log

### [Date will be added as work happens]
- Work completed
- Decisions made
- Notes for next session

---

## Files Changed
(Will be updated as implementation progresses)

## Architectural Decisions
(Major technical decisions and rationale)

## Lessons Learned
(What worked, what didn't, what to do differently)

Step 6: Next Step

After creating progress doc:

"Progress tracking set up! Document saved to docs/{name}/PROGRESS.md

Next step: Run /build phase 1 {name} to begin implementation."


Subcommand: phase

Step 1: Parse Arguments

Parse arguments to extract:

  • Phase number (if provided)
  • Feature name (if provided)

If neither provided, prompt for both using AskUserQuestion.

Step 2: Get Feature Name

If feature name not determined, use AskUserQuestion to prompt for it.

Step 3: Verify All Docs Exist

Check that all three docs exist:

  • docs/{name}/RESEARCH.md
  • docs/{name}/IMPLEMENTATION.md
  • docs/{name}/PROGRESS.md

If any missing, inform user which doc is missing and suggest the appropriate /build command to create it.

Step 4: Get Phase Number

If phase number not in arguments:

Read docs/{name}/IMPLEMENTATION.md to extract available phases.

Use AskUserQuestion to let user select:

  • question: "Which phase would you like to work on?"
  • header: "Phase"
  • multiSelect: false
  • options: [dynamically generated from phases found in IMPLEMENTATION.md, marking completed ones]

Step 5: Read All Context

Read all three documents to fully understand:

  • The research and rationale (RESEARCH.md)
  • The specific phase tasks and success criteria (IMPLEMENTATION.md)
  • Current progress and decisions made (PROGRESS.md)

Step 6: Deep Research on Phase

Before starting implementation:

  1. Web search for specific implementation details relevant to this phase
  2. Codebase exploration for relevant existing code
  3. Use AskUserQuestion to clarify any ambiguities about the phase requirements

Step 7: Execute Phase Work

Begin implementing the phase:

  1. Work through each task in the phase
  2. Use AskUserQuestion frequently for implementation decisions
  3. Follow the "Always Works" philosophy - test as you go
  4. Document decisions in PROGRESS.md as you make them

Step 8: Update Progress Document

As you work, update docs/{name}/PROGRESS.md:

  • Mark tasks as completed
  • Record decisions made and why
  • Note any blockers encountered
  • List files changed
  • Add architectural decisions
  • Update the session log with today's work

Update the phase status:

  • "In Progress" when starting
  • "Completed" when all tasks done and success criteria met

Step 9: Next Step

After completing the phase:

  1. Read PROGRESS.md to determine next incomplete phase
  2. Inform user of completion and suggest next action:

"Phase {n} complete! Progress updated in docs/{name}/PROGRESS.md

Next step: Run /build phase {n+1} {name} to continue with [next phase title]."

Or if all phases complete:

"All phases complete! The {feature name} feature implementation is done.

Consider:

  • Running tests to verify everything works
  • Updating documentation
  • Creating a PR for review"

Subcommand: status

Step 1: Get Feature Name

If feature name not in arguments, use AskUserQuestion to prompt for it.

Step 2: Check Which Docs Exist

Check for existence of:

  • docs/{name}/RESEARCH.md
  • docs/{name}/IMPLEMENTATION.md
  • docs/{name}/PROGRESS.md

Step 3: Determine Status and Next Step

Based on which docs exist:

No docs exist: "No documents found for feature '{name}'. Next step: Run /build research {name} to start."

Only RESEARCH.md exists: "Research complete for '{name}'. Next step: Run /build implementation {name} to create implementation plan."

RESEARCH.md and IMPLEMENTATION.md exist: "Research and implementation plan complete for '{name}'. Next step: Run /build progress {name} to set up progress tracking."

All three exist: Read PROGRESS.md to find current phase status. "Feature '{name}' is in progress. Current status: [Phase X - status] Next step: Run /build phase {next incomplete phase} {name} to continue."

If all phases complete: "Feature '{name}' implementation is complete!"


Important Guidelines

Use AskUserQuestion Liberally

Throughout all phases, use AskUserQuestion whenever:

  • There's ambiguity in requirements
  • Multiple approaches are possible
  • You need to validate an assumption
  • A decision will significantly impact the implementation
  • You're unsure about scope or priority

Deep Research Expectations

"Deep research" means:

  • Multiple web searches on different aspects
  • Thorough codebase exploration
  • Reading relevant documentation
  • Considering multiple approaches
  • Understanding trade-offs

Don't rush through research - it's the foundation for good implementation.

Progress Tracking

Keep PROGRESS.md updated in real-time during phase work:

  • Don't wait until the end to update
  • Record decisions as they're made
  • Note blockers immediately
  • This creates valuable context for future sessions

Scope Management

A key purpose of this workflow is preventing scope creep:

  • Each phase should have clear boundaries
  • If new requirements emerge, note them for future phases
  • Don't expand the current phase's scope mid-implementation
  • Use AskUserQuestion to validate if something is in/out of scope

Always Works Philosophy

When implementing phases:

  • Test changes as you make them
  • Don't assume code works - verify it
  • If something doesn't work, fix it before moving on
  • The goal is working software, not just written code

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

37.99%
按下载量换算141

Claude

27.17%
按下载量换算101

Cursor

18.36%
按下载量换算68

Gemini CLI

10.38%
按下载量换算39

安全审计

暂无安全审计结果可展示。

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills