Token导航 LogoToken导航

Agent Skill · 研究检索

client-package-local(客户端包本地)

client-package-local 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。

client-package-local 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,权限为调用外部服务;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 68,884。

总安装 222周安装 9GitHub Stars 607下载 70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:client-package-local(客户端包本地)
来源仓库:https://github.com/athina-ai/goose-skills
仓库路径:skills/client-package-local
安装命令:
npx skills add https://github.com/athina-ai/goose-skills --skill client-package-local
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/athina-ai/goose-skills --skill client-package-local

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。

client-package-local 用于将客户工作整合为结构化的本地交付包,包含按日期命名的 .md 文件和 Google Sheets。适合在 Codex、Claude、Cursor、Gemini CLI 中快速构建可导航的成果目录,适用于项目交付和文档整理场景。

通过指定客户名称、日期和介绍信息生成文件,支持自定义收件人和上下文背景。安装前需确认仓库权限、维护状态及是否涉及文件读写或网络访问。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/client-package-local

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/athina-ai/goose-skills --skill client-package-…;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

    它需要调用外部服务,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

client-package-local 怎么安装?

在终端执行 npx skills add https://github.com/athina-ai/goose-skills --skill client-package-local。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

client-package-local 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。

client-package-local 需要什么权限?

权限级别是调用外部服务。该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

client-package-local 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。

client-package-local 用的人多吗?

公开数据是累计安装 222 次,近一周安装 9 次,下载 70 次,GitHub 607 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。

client-package-local 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

  • 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

  • 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
client-package-local

客户端包本地 · 本页

222607调用外部服务未明确通过

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

create-client-package

Package all GTM work for a client into a structured local delivery package with dated.md files and Google Sheets. Reads the client's workspace folder and builds a navigable directory of deliverables.

Inputs

InputRequiredDefaultDescription
client_nameYes--Client folder name under clients/ (e.g., truewind)
dateNotoday's dateDate string in YYYY-MM-DD format, used for folder and file naming
intro_messageNo--Custom introduction message for the overview file. If not provided, generate one based on assets found.
recipient_nameNo--Name of the person receiving the package (used in intro)
recipient_contextNo--Any framing context for the delivery (e.g., "we built these to capitalize on the Botkeeper shutdown")
include_strategiesNotrueWhether to include strategy documents
include_campaignsNotrueWhether to include campaign assets
include_contentNotrueWhether to include content drafts
include_leadsNotrueWhether to include lead lists (uploaded to Google Sheets)

Setup

Requires access to:

  • Rube MCP server — for creating Google Sheets (via Composio GOOGLESHEETS tools)

No API keys need to be set manually — Google Sheets access is through MCP.

Output Directory Structure

clients/<client_name>/client-package/<YYYY-MM-DD>/
├── Overview - <YYYY-MM-DD>.md
├── Lead Lists - <YYYY-MM-DD>.md
└── Strategies - <YYYY-MM-DD>/
    ├── <Strategy 1 Name>/
    │   ├── overview.md
    │   ├── <Substrategy 1 Name> - <YYYY-MM-DD>.md
    │   ├── <Substrategy 2 Name> - <YYYY-MM-DD>.md
    │   ├── <substrategy-1-leads>.csv
    │   └── <substrategy-2-leads>.csv
    ├── <Strategy 2 Name>/
    │   ├── overview.md
    │   ├── ...
    └── ...

Procedure

Step 1: Scan the Client Folder

Read the client folder at clients/<client_name>/ and inventory all available assets:

clients/<client_name>/
├── context.md          # Client context, ICP, positioning
├── competitors.md      # Competitive landscape (optional)
├── notes.md            # Running log of decisions
├── strategies/         # Strategy documents (*.md)
├── campaigns/          # Campaign assets (folders or *.md)
├── content/            # Content drafts (blog posts, comparison pages, etc.)
└── leads/              # Lead lists (*.csv, *.json, *.md)

For each directory, list all files and read their contents. Build an inventory:

  • Strategies: List of .md files in strategies/ (skip ORCHESTRATION-PROMPT.md and other internal-only files)
  • Campaigns: List of campaign folders or .md files in campaigns/
  • Content: List of .md files in content/
  • Leads: List of .csv and .json files in leads/ (for Google Sheets upload) and .md files (for reference)

If a directory doesn't exist or is empty, skip it.

Step 2: Identify Strategies and Map Assets

Each strategy in strategies/ is a top-level theme. For each strategy:

  1. Read the strategy.md file to understand the strategy name, summary, and execution plan
  2. Map campaigns to strategies — match campaigns in campaigns/ to strategies by name/theme (e.g., hiring-hurting-outreach maps to the hiring-hurting strategy)
  3. Map content to strategies — match content in content/ to strategies by name/theme (e.g., botkeeper-shut-down-what-accounting-firms-should-do-next.md maps to botkeeper-shutdown strategy)
  4. Map leads to strategies — match lead files in leads/ to strategies by name/theme (e.g., botkeeper-linkedin-leads.csv maps to botkeeper-shutdown strategy)

Assets that don't clearly map to a strategy should be grouped under a "General" or "Ungrouped" section.

Step 3: Upload Lead Lists to Google Sheets

For each lead list file (.csv or .json) found in leads/:

  1. Parse the file and extract structured data

- For .csv: parse directly - For .json: read the JSON, flatten into tabular format

  1. Ensure required columns exist. Every Google Sheet MUST include these columns (add them if missing, populate with available data or mark as "N/A"): Additional columns from the source data should be preserved after the required columns.

- Name — lead's full name - Company — company name - Title — job title - LinkedIn URL — LinkedIn profile URL - Source — how we found them (e.g., "LinkedIn Post", "Job Posting", "Conference", "Web Archive") - Qualification Status — e.g., "Qualified", "Not Qualified", "Needs Review" - Qualification Reasoning — why they qualified or didn't

  1. Create a new Google Sheet using RUBE_SEARCH_TOOLS to find GOOGLESHEETS_CREATE_GOOGLE_SHEET1, then execute:

- Title format: <Client Name> — <Lead List Name> (<date>) - Example: Truewind — Botkeeper LinkedIn Leads (2026-02-24)

  1. Write data using GOOGLESHEETS_BATCH_UPDATE:

- First row = headers (required columns first, then additional columns) - Remaining rows = data - Use first_cell_location: "A1" and valueInputOption: "USER_ENTERED"

  1. Record the spreadsheet URL for linking in the Lead Lists summary file

If there are multiple lead files, create sheets in parallel where possible.

Step 4: Create the Package Directory

Create the output directory structure:

mkdir -p clients/<client_name>/client-package/<date>
mkdir -p clients/<client_name>/client-package/<date>/Strategies - <date>

Step 5: Create the Overview File

Create Overview - <date>.md with:

  • A greeting to the recipient (if recipient_name provided)
  • A summary of the engagement — what strategies were developed, how many leads were found, what campaigns were built
  • A table of contents linking to the other files in the package
  • Any framing context from recipient_context
  • A closing line

Example structure:

# GTM Engineering Package — <Client Name>

**Prepared:** <date>
**For:** <recipient_name> (if provided)

## Summary

<Brief overview of what was done — strategies developed, leads found, campaigns built>

## What's Inside

### Strategies
- **<Strategy 1>** — <one-line summary>
- **<Strategy 2>** — <one-line summary>

### Lead Lists
See [Lead Lists - <date>](./Lead Lists - <date>.md) for all lead lists with Google Sheet links.

**Total leads found:** <count across all lists>

### Campaigns
<List of campaigns built, with which strategy they belong to>

### Content
<List of content pieces created>

---

<closing line>

Step 6: Create the Lead Lists File

Create Lead Lists - <date>.md with:

  • A bulleted list of all Google Sheet links created in Step 3
  • For each sheet: the name, number of leads, a brief description of the source/strategy, and the Google Sheet URL

Example:

# Lead Lists — <Client Name>

**Date:** <date>

## Sheets

- **[Truewind — Botkeeper LinkedIn Leads](https://docs.google.com/spreadsheets/d/...)** — 10 leads from LinkedIn posts/comments about Botkeeper shutdown
- **[Truewind — Hiring Signal Leads](https://docs.google.com/spreadsheets/d/...)** — 25 leads from job posting intent signals
- **[Truewind — Conference Speakers Q1-Q2 2026](https://docs.google.com/spreadsheets/d/...)** — 15 leads from accounting conference speaker lists

**Total:** X leads across Y sheets

Step 7: Create Strategy Subfolders

For each strategy identified in Step 2, create a subfolder under Strategies - <date>/:

Strategies - <date>/
└── <Strategy Name>/
    ├── overview.md
    ├── <Substrategy/Campaign 1> - <date>.md
    ├── <Substrategy/Campaign 2> - <date>.md
    ├── <substrategy-1-leads>.csv
    └── <substrategy-2-leads>.csv
overview.md

A summary of the strategy:

  • Strategy name and one-paragraph summary
  • The signal being tracked
  • Target ICP / filters
  • List of substrategies, campaigns, and content pieces with brief descriptions
  • Links to the Google Sheets for any related lead lists
Substrategy / Campaign.md files

For each campaign or content asset mapped to this strategy:

  • Copy the meaningful content from the original file
  • Clean up any internal-only notes or orchestration details
  • Include links to relevant Google Sheets
  • Name the file: <Descriptive Name> - <date>.md
Lead.csv files

For each lead list that maps to this strategy:

  • Export a copy of the lead data as a .csv file in the strategy subfolder
  • Use the standardized columns (Name, Company, Title, LinkedIn URL, Source, Qualification Status, Qualification Reasoning) plus any additional relevant columns
  • Name the file descriptively: <strategy-name>-leads.csv or <specific-source>-leads.csv

Step 8: Verify and Report

After all files are created, output a summary:

## Package Created

**Location:** clients/<client_name>/client-package/<date>/

**Files:**
- Overview - <date>.md
- Lead Lists - <date>.md
- Strategies - <date>/
  - <Strategy 1>/
    - overview.md
    - <Substrategy 1> - <date>.md
    - <substrategy-1>-leads.csv
  - <Strategy 2>/
    - overview.md
    - ...

**Google Sheets:**
- [Lead List 1](<sheets-url>) — 12 leads
- [Lead List 2](<sheets-url>) — 9 leads

**Total:** X files created, Y Google Sheets created, Z total leads

Example Prompts

  • "Package everything we've done for Truewind into a shareable folder"
  • "Create a client delivery package for Acme Corp with all our work"
  • "Build a package for [client] — include strategies, campaigns, and lead lists"
  • "Take our work in clients/truewind and create a deliverable"
  • "Package the client folder into something I can send to the founder"

Troubleshooting

ProblemSolution
Google Sheets creation failsVerify the Rube MCP server is connected and Google Sheets has an active connection. Run RUBE_MANAGE_CONNECTIONS with toolkits: ["googlesheets"].
CSV parsing errorsCheck the CSV for encoding issues. The skill expects UTF-8 CSVs with a header row.
JSON lead files have nested structureFlatten the JSON to tabular format before uploading. Extract the key fields into the required columns.
Lead list is empty or malformedSkip that file and note it in the summary. Don't create an empty Google Sheet.
Strategy mapping is ambiguousWhen a campaign or lead list could belong to multiple strategies, place it under the most specific match. If truly ambiguous, ask the user.
No strategies folder existsCreate a single "General" strategy folder and place all assets there.

Metadata

metadata:
  requires:
    mcp_servers: ["rube"]
  cost: "Free (Google Sheets via Composio)"

平台分布

Codex

34.31%
按下载量换算24

Claude

28.47%
按下载量换算20

Cursor

18.56%
按下载量换算13

Gemini CLI

9.72%
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

调用外部服务

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

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