Token导航 LogoToken导航

Agent Skill · 研究检索

developer-summary(开发者总结)

developer-summary 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。

developer-summary 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,权限为接触敏感数据;许可证信息还不明确,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 69,087。

总安装 220周安装 9GitHub Stars 公开资料未说明下载 71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
unknown

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:developer-summary(开发者总结)
来源仓库:https://github.com/validkeys/sherpy
仓库路径:skills/developer-summary
安装命令:
npx skills add https://github.com/validkeys/sherpy --skill developer-summary
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/validkeys/sherpy --skill developer-summary

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。

developer-summary 根据 Sherpy 规划工件自动生成开发者视角的摘要文档,突出关键交付物和里程碑。自动识别 requirements/business-requirements.yaml 所在目录,若无则引导用户提供路径。输出包含项目概览、核心功能列表和时间线视图,便于新成员快速掌握项目轮廓。

依赖标准文件夹结构中的 artifacts 文件,若缺少必要工件将中断生成流程并提示补全。适用于项目启动阶段或交接场景,辅助建立统一认知基准而非替代详细规格说明。

来源仓库内路径:skills/developer-summary

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/validkeys/sherpy --skill developer-summary;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

    它需要接触敏感数据,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

developer-summary 怎么安装?

在终端执行 npx skills add https://github.com/validkeys/sherpy --skill developer-summary。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

developer-summary 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。

developer-summary 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

developer-summary 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

developer-summary 用的人多吗?

公开数据是累计安装 220 次,近一周安装 9 次,下载 71 次。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。

developer-summary 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。
  • 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

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 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。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要明确授权才能用到公司项目里的场景。
  • 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
developer-summary

开发者总结 · 本页

220—接触敏感数据未明确通过

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

Developer Summary

Generates a developer-focused summary document from Sherpy planning artifacts. Provides developers with a quick-start view of what's being built, key deliverables, and the milestone roadmap.

Usage

/developer-summary [base-directory]

If no directory is provided, auto-detect by looking for requirements/business-requirements.yaml in the current directory.

If not found, prompt the user: "Where are your planning artifacts located?"

Wait for the user to provide a path before proceeding.

Required Artifacts

The skill auto-discovers these files from the standard Sherpy folder structure:

FileLocationPurpose
business-requirements.yamlrequirements/Project overview and deliverables
technical-requirements.yamlrequirements/Technical context and architecture
milestones.yamlimplementation/Milestone sequence and estimates
timeline.yamldelivery/Timeline estimates and dates

Process

Step 1: Determine Base Directory and Scan for Required Artifacts

If no directory parameter was provided, check if requirements/business-requirements.yaml exists in the current directory.

  • If found, use current directory as base_directory
  • If not found, prompt: "Where are your planning artifacts located?" and wait for user response

Once base_directory is determined, scan for required files in standard locations. Display status:

## Developer Summary — Dependency Check

Scanning for required artifacts...

 ✓  requirements/business-requirements.yaml
 ✓  requirements/technical-requirements.yaml
 ✓  implementation/milestones.yaml
 ✗  delivery/timeline.yaml

3 of 4 required files found.

Legend: ✓ found · ✗ missing

Step 2: Handle Missing Files

If any required files are missing, ask:

"Missing required files: timeline.yaml Options: 1. Continue with partial data — Generate summary with available information (missing sections will be marked as [Not Available]) 2. Exit — Complete missing artifacts first (run /delivery-timeline to generate timeline.yaml) What would you like to do?"

Wait for user response:

  • If user chooses 1 (Continue): Proceed with available data, mark missing sections clearly
  • If user chooses 2 (Exit): Display which skills to run and exit gracefully

Step 3: Load and Parse Artifacts

Read all available files and extract:

From business-requirements.yaml:

  • Project name and description
  • Core features and capabilities
  • Key deliverables

From technical-requirements.yaml:

  • Technology stack
  • Architectural patterns
  • Key technical components

From milestones.yaml:

  • Milestone sequence (m0, m1, m2, etc.)
  • Milestone names and descriptions
  • Dependencies between milestones
  • Estimated duration per milestone

From timeline.yaml (if available):

  • Total project duration
  • Target completion date
  • Timeline workback schedule

Step 4: Generate Developer Summary

Create summaries/developer-summary.md.

The output document includes: header (timestamp, project name), Overview (3-5 sentence synthesis), Deliverables (grouped by category), Milestones & Timeline (M0, M1... with duration, dependencies, key deliverable), and Summary (total milestones, estimated duration, target completion).

See references/output-spec.md for the complete document specification with section structure and formatting rules.

See references/example.md for a full example.

Step 5: Handle Missing Data Gracefully

When required files are missing, insert clear placeholders:

## Milestones & Timeline

[Not Available - `milestones.yaml` not found. Run `/implementation-planner` to generate milestones.]

Or for partial data:

## Summary

**Total Milestones:** 4
**Estimated Total Duration:** 6 weeks (estimated from milestones)
**Target Completion:** [Not Available - run `/delivery-timeline` to set production deploy date]

Step 6: Confirmation

After generating the summary, display:

## Developer Summary Generated ✓

**Location:** summaries/developer-summary.md
**Size:** [file size]

Summary includes:
  ✓ Project overview
  ✓ Deliverables breakdown
  ✓ Milestone roadmap
  [Missing data: timeline completion date - run /delivery-timeline]

The summary is ready for developer onboarding and reference.

Deliverables Extraction Logic

Identifying Deliverables

Parse the following sources to build the deliverables list:

From business-requirements.yaml:

  • functional_requirements → User-facing features
  • features or capabilities → High-level capabilities
  • scope.in_scope → Explicitly included items

From technical-requirements.yaml:

  • architecture.components → Technical components
  • technology_stack → Infrastructure items (databases, APIs, services)
  • data_model.entities → Domain entities/models
  • apis.endpoints → API endpoints if listed
  • integration_points → External integrations

Categorization Rules

Group deliverables into logical categories based on type:

  • API Endpoints - REST/GraphQL endpoints, webhooks
  • Domain Handlers - Business logic handlers, command processors
  • UI Components - Pages, views, components (if applicable)
  • Data Models - Entities, schemas, database tables
  • Infrastructure - Databases, message queues, auth systems
  • Integrations - External service connections
  • Background Jobs - Scheduled tasks, workers
  • Configuration - Environment setup, deployment configs

Use the most relevant categories based on what's present in the requirements. Don't force categories that don't apply.

Overview Generation Logic

Crafting the Overview Paragraph

Synthesize information from multiple sources to create a concise, informative overview:

Elements to include (in order):

  1. What - The system/feature being built (1 sentence)
  2. Why - The problem it solves or value it provides (1 sentence)
  3. How (high-level) - The approach or key capability (1 sentence, optional)

Source fields to use:

  • business-requirements.yaml:

- description or overview - business_goals.primary_goal - success_criteria

  • technical-requirements.yaml:

- solution_overview - architecture.approach

Template:

"[System Name] is a [type of system] that [primary function]. It addresses [problem] by [solution approach], enabling [key benefit/outcome]. The system will [key technical approach or differentiator]."

Example:

"The Sherpy CLI is a structured planning tool for AI-assisted software development. It addresses requirement drift and planning gaps by conducting guided interviews and generating comprehensive implementation plans with task-level detail. The system enforces best practices like task sizing, TDD requirements, and style anchors throughout the planning process."

Keep it Concise

  • Maximum 5 sentences
  • No bullet points in the overview
  • Focus on business value, not implementation details
  • Avoid jargon unless it's domain-specific and necessary

Milestone Roadmap Logic

Extracting Milestone Information

For each milestone in milestones.yaml, extract:

Required fields:

  • id - Milestone identifier (m0, m1, etc.)
  • name - Milestone title
  • description - What the milestone delivers
  • estimated_duration - Time estimate
  • dependencies - Which milestones must complete first

Formatting:

  • Display milestones in sequence (m0 → m1 → m2...)
  • Show dependencies clearly
  • Highlight the "key deliverable" from the first line of the description or the most important success criterion

Example extraction:

# From milestones.yaml:
milestones:
  - id: m1
    name: API Foundation
    description: |
      Build core REST API with authentication and authorization.
      Includes user management endpoints and JWT token handling.
    dependencies: [m0]
    estimated_duration: 2 weeks
    success_criteria:
      - All auth endpoints functional
      - JWT tokens properly validated

Renders as:

### M1: API Foundation
**Duration:** 2 weeks
**Dependencies:** M0
**Key Deliverable:** Core REST API with authentication and authorization

Timeline Calculation

If timeline.yaml is available:

  • Use total_duration field
  • Use production_deploy_date for target completion
  • Show the calculated project dates

If only milestones.yaml is available:

  • Sum all estimated_duration values
  • Convert to appropriate unit (days/weeks)
  • Note that this is an estimate without QA/PR time

If neither is available:

  • Mark as "TBD" with instruction to run appropriate skill

Error Handling

File Not Found

If a required file doesn't exist in the expected location:

  • Check alternative common locations (root directory as fallback)
  • Display clear error with exact path tried
  • Suggest which Sherpy skill generates that file

Invalid YAML

If a file exists but can't be parsed:

  • Display parsing error
  • Show the file path and line number if available
  • Suggest checking file format or regenerating

Missing Required Fields

If a file is missing expected fields:

  • Use sensible defaults or mark as "[Not Specified]"
  • Continue generating summary with available data
  • Note missing fields in the confirmation output

Output Location

Always output to: {base_directory}/summaries/developer-summary.md

If the directory doesn't exist, create it:

mkdir -p {base_directory}/summaries

Integration with Sherpy Flow

This skill is designed to be called as Step 10 in /sherpy-flow. When invoked by sherpy-flow, it receives base_directory as a parameter. It should:

  • Run after all planning artifacts are complete
  • Auto-discover files from the organized folder structure within base_directory
  • Not fail if optional files (like timeline) are missing
  • Generate the summary without user interaction when all files are present

Examples

See references/example.md for a complete sample developer summary.

Related Skills

  • /executive-summary - Generate executive summary for stakeholders
  • /sherpy-flow - Full planning workflow (includes this skill at the end)
  • /implementation-planner - Generates the milestones.yaml used by this skill
  • /delivery-timeline - Generates the timeline.yaml used by this skill

平台分布

Codex

33.71%
按下载量换算24

Claude

31.26%
按下载量换算22

Cursor

18.54%
按下载量换算13

Gemini CLI

8.62%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills