Token导航 LogoToken导航TokenDH.com
前端设计执行命令github未标认证来源可访问许可证需确认审计通过

backlogbacklog 搜索

Agent Skill

backlog 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。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。

总安装

360

周安装

15

GitHub Stars

公开资料未说明

下载量

120
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:backlog(backlog 搜索)
来源仓库:https://github.com/miguelalcalde/agentfiles
仓库路径:skills/backlog
安装命令:
npx skills add https://github.com/miguelalcalde/agentfiles --skill backlog
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/miguelalcalde/agentfiles --skill backlog

简介

backlog 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息。

  • 适合在 Codex、Claude、Cursor、Gemini CLI 中围绕仓库状态、代码变更或协作事项进行整理。
  • 通过 npx skills add 命令从指定仓库安装并使用该技能。
  • 安装前需确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。
  • backlog 属于前端设计类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Backlog Workflow

A structured workflow for taking features from backlog to implementation using specialized agents.

Workflow Overview

Backlog → Triage → PRD → Plan → Refine → PRD (refined) → Implement → Code

Manual mode: Each step is human-triggered. Agents do not auto-chain.

Conductor mode: Run the conductor workflow to orchestrate all phases automatically until complete or blocked.

Workflow Actions

ActionAgentPurposeOptional command alias
TriagetriagerSelect task from backlog, create blank PRD/triage
Plan [slug]plannerCreate implementation plan/plan [slug]
Refine [slug]refinerComplete and validate PRD/refine [slug]
Implement [slug]implementerExecute plan on feature branch/implement [slug]
ConductconductorOrchestrate all phases in a loop/conduct
Conduct phased runconductorRun specific phases only (e.g., triage,plan)/conduct --phases X
Conduct targeted runconductorProcess specific feature only/conduct --slug X

Slash commands are optional convenience wrappers. The workflow is designed to work with plain prompts and subagent invocation across tools.

Agents

Agents are minimal — they define role, boundaries, and tools. Methodology lives in skills.

AgentBranchWrites ToMethodology Skill
triagermain.backlog/prds/, .backlog/backlog.mdskills/backlog/triage/SKILL.md
refinermain.backlog/prds/skills/backlog/refine/SKILL.md
plannermain.backlog/plans/, .backlog/prds/skills/backlog/plan/SKILL.md
implementerfeature/*Source codeskills/backlog/implement/SKILL.md
conductormain + feat.backlog/, source code, action logOrchestrates all of the above

Naming Convention

The workflow uses descriptive slugs instead of numeric IDs:

ArtifactFormatExample
Backlog entry[slug] Title[user-auth] User Authentication
PRD filePRD-[slug].mdPRD-user-auth.md
Plan filePLAN-[slug].mdPLAN-user-auth.md
Branchfeat/[slug]feat/user-auth

Slugs: lowercase kebab-case, max 30 characters.

Status Flow

PRD Statuses

blank → refined → needs_review → approved

Only a human sets approved. The refiner sets refined or needs_review.

Plan Statuses

draft → needs_review → approved → implemented | partially_implemented

Only a human sets approved. The planner sets draft or needs_review.

Human Checkpoints

  • After Triage: Review selected task, adjust if needed
  • After Plan: Review plan, mark as approved if ready
  • After Refine: Review PRD, mark as approved if ready
  • After Implement: Review code, create PR manually

Project Setup

Each project using this workflow needs a .backlog/ directory:

your-project/
└── .backlog/
    ├── backlog.md
    ├── prds/
    └── plans/

Example Usage

Manual Mode (step-by-step)

# Triage the highest priority task
"Use the triager agent to pick the highest-priority pending task and create a PRD."

# Create implementation plan
"Use the planner agent to create the implementation plan for user-auth."

# Refine a specific PRD
"Use the refiner agent to refine PRD-user-auth."

# Execute the plan
"Use the implementer agent to implement the approved plan for user-auth."

Conductor Mode (automated)

# Run full pipeline until complete or blocked
"Use the conductor agent to run the backlog pipeline until complete or blocked."

# Run only triage and plan phases
"Use the conductor agent and run only phases: triage,plan."

# Process specific feature only
"Use the conductor agent and process only slug: user-auth."

# Named conductor for parallel operation
"Use conductor name frontend and run phases triage,plan."

Parallel Conductors

Multiple conductors can run in parallel when handling different phases:

# Terminal 1: Create PRDs and plans
"Use conductor name frontend and run phases triage,plan."

# Terminal 2: Review plans
"Use conductor name reviewer and run phase refine."

# Terminal 3: Implement approved plans
"Use conductor name builder and run phase implement."

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

能力 4

展示第三方安全扫描或审计结果

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

36.49%
按下载量换算44

Claude

29.6%
按下载量换算36

Cursor

18%
按下载量换算22

Gemini CLI

9.5%
按下载量换算11

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

执行命令

安装流程涉及命令执行,可能通过 npx skills add https://github.com/miguelalcalde/agentfiles --skill backlog 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

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