Token导航 LogoToken导航

Agent Skill · 研究检索

client-packet-engine(客户端数据包引擎)

client-packet-engine 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-packet-engine 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,第三方扫描有提醒项,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 607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:client-packet-engine(客户端数据包引擎)
来源仓库:https://github.com/athina-ai/goose-skills
仓库路径:skills/client-packet-engine
安装命令:
npx skills add https://github.com/athina-ai/goose-skills --skill client-packet-engine
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-packet-engine

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-packet-engine 用于批量生成面向潜在客户的 GTM 提案数据包,包含情报资料、增长策略和示例执行内容。适合在 Codex、Claude、Cursor、Gemini CLI 中为多公司准备营销材料,适用于销售前期支持。自动输出潜在客户列表、内容草稿和外呼序列,但不执行实际投放活动。

安装前建议核实数据源权限、输出范围及是否会触发外部请求或文件操作。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/client-packet-engine

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-packet-e…;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-packet-engine 怎么安装?

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

client-packet-engine 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-packet-engine 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-packet-engine 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-packet-engine 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,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-packet-engine 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

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。

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。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 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-packet-engine

客户端数据包引擎 · 本页

220607调用外部服务未明确有提醒

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

Client Packet Engine

Batch pipeline that takes a list of companies and produces finished GTM pitch packets — intelligence packages, growth strategies, and executed sample work (lead lists, content drafts, email sequences). Designed for prospective client pitches, not live campaign execution.

When to Use

  • "Build pitch packets for [company list]"
  • "Create client packets for these companies"
  • "Run the full pitch pipeline for [companies]"
  • "Prepare GTM packets for prospective clients"

Key Constraints

  • Pitch-packet mode: Generates sample leads, content, and outreach drafts but does NOT launch live campaigns, send real emails, or spend on paid Apollo enrichment. The goal is demonstrating value.
  • Cost budget: Default $5/company. Warn at 80% ($4). Hard stop at limit.
  • Parallelization: Up to 3 companies concurrent (Phase 1), up to 5 strategy executions concurrent (Phase 3).
  • Human checkpoints: (1) After cost estimate, (2) Strategy selection per company, (3) Before any single API call >$2.

Input

companies:
  - name: "Company Name"
    url: "https://company.com"
  - name: "Another Company"
    url: "https://another.com"

Minimum: 1 company. No hard maximum, but warn if >10 (long runtime, high cost).


Phase 0: Intake & Validation

Steps

  1. Parse input: Extract company names and URLs from the user's request. Normalize URLs (add https:// if missing, strip trailing slashes).
  2. Validate websites: For each URL, do a quick web fetch to confirm:

- Site is reachable (not 404/500) - It's the correct company (page title or content matches company name) - Flag any mismatches for user confirmation

  1. Estimate costs: Calculate per-company estimate based on typical skill costs:

- Phase 1 (research + audits): ~$1.50/company (Apify credits for SEO, ads, content scraping) - Phase 3 (strategy execution): ~$1.00-3.00/company depending on strategy count and types - Total estimate: $2.50-4.50/company typical

  1. Present summary for approval: Batch Summary: ┌──────────────────┬─────────────────────────┬──────────┐ │ Company │ URL │ Est Cost │ ├──────────────────┼─────────────────────────┼──────────┤ │ Acme Corp │ https://acme.com │ ~$3.50 │ │ Beta Inc │ https://beta.io │ ~$3.50 │ └──────────────────┴─────────────────────────┴──────────┘ Total estimated cost: ~$7.00 Budget limit: $5.00/company ($10.00 total)
  2. Get user approval before proceeding.

Output

  • Validated company list with confirmed URLs
  • Cost estimate acknowledged by user

Phase 1: Research & Strategy Generation

Run the client-onboarding playbook (Phases 1-3) for each company.

Execution

  • Process up to 3 companies in parallel using Task agents
  • Each company gets the full client-onboarding treatment:

- Phase 1: Intelligence Gathering (company research, competitor research, founder research, SEO audit, AEO visibility, ads review, industry scan, GTM analysis) - Phase 2: Synthesis into Client Intelligence Package - Phase 3: Strategy Generation with execution tags (see Structured Execution Tags in client-onboarding)

Cost Tracking

After each company completes:

  • Log actual Apify/API costs
  • Compare against per-company budget
  • If at 80% of budget, warn before continuing
  • If at limit, stop and report what was completed

Output per Company

  • clients/<company>/intelligence-package.md
  • clients/<company>/growth-strategies.md (with <!-- execution --> tags on each strategy)
  • All Phase 1 research files in clients/<company>/intelligence/

Phase 2: Strategy Selection (Human Checkpoint)

Present all generated strategies across all companies for user selection.

Presentation Format

For each company, show a summary table:

## Acme Corp — 7 strategies generated

| # | Strategy | Pattern | Priority | ICE | Est Leads | Est Cost |
|---|----------|---------|----------|-----|-----------|----------|
| 1 | Hiring signal outbound (DevOps roles) | signal-outbound | P0 | 8.2 | ~50 | $0.80 |
| 2 | Competitor displacement (vs BigCo) | competitive-displacement | P0 | 7.5 | ~30 | $1.20 |
| 3 | Conference speaker prospecting | event-prospecting | P1 | 6.8 | ~20 | $0.50 |
| 4 | SEO comparison content | content-lead-gen | P1 | 6.5 | ~40 | $0.60 |
| 5 | Quarterly business review timing | lifecycle-timing | P2 | 5.0 | ~15 | $0.30 |
| 6 | LinkedIn thought leadership | manual | P2 | 4.5 | — | — |
| 7 | Partner co-marketing | manual | P2 | 3.8 | — | — |

Select strategies to execute: [e.g., "1-4", "all P0", "top 3 by ICE"]

Selection Options

Accept flexible input:

  • Specific numbers: "1, 3, 5"
  • Ranges: "1-4"
  • Priority-based: "all P0", "P0 and P1"
  • Score-based: "top 3 by ICE"
  • All executable: "all" (skips manual pattern strategies)
  • Per-company overrides: "Acme: 1-3, Beta: all P0"

Validation

  • Warn if selected strategies exceed remaining per-company budget
  • Flag any strategies tagged manual (these produce a plan document but no automated execution)
  • Confirm final selection before proceeding

Phase 3: Strategy Execution (Pitch-Packet Mode)

Execute selected strategies using the execution pattern routing system.

Execution Pattern Router

Read the <!-- execution --> tag from each selected strategy and route to the appropriate skill chain.

Pattern: signal-outbound

Detect buying signal → find contacts → draft outreach.

StepSkillPitch-Packet Behavior
1. Detect signalsjob-posting-intent / linkedin-post-research / review-scraper (based on signal_type)Run normally — free or low-cost scraping
2. Find contactscompany-contact-finderRun normally — uses Gooseworks MCP (free)
3. Draft outreachemail-draftingDraft sequences but do NOT send or set up campaigns

Output: Signal report + lead CSV + draft email sequences

Pattern: content-lead-gen

Create content asset + find leads who'd care.

StepSkillPitch-Packet Behavior
1. Create contentcontent-asset-creatorGenerate the asset (HTML report, comparison page, etc.)
2. Find leadsapollo-lead-finderFree search only — no paid enrichment. Show lead count + titles

Output: Content asset (HTML) + lead list (names/titles/companies, no emails)

Pattern: competitive-displacement

Target competitor's unhappy customers.

StepSkillPitch-Packet Behavior
1. Mine competitor intelweb-archive-scraper + review-scraperScrape archived customer lists + negative reviews
2. Find contactscompany-contact-finderFind decision-makers at identified companies
3. Draft outreachemail-draftingDraft displacement-themed sequences
4. Create comparison contentcontent-asset-creatorBuild comparison page (client vs competitor)

Output: Competitor intel report + lead CSV + email sequences + comparison page

Pattern: event-prospecting

Find event attendees/speakers matching ICP.

StepSkillPitch-Packet Behavior
1. Find events/peopleluma-event-attendees / conference-speaker-scraperScrape relevant events
2. Qualify leadslead-qualificationScore against ICP
3. Find contactscompany-contact-finderEnrich qualified leads

Output: Event attendee list + qualified leads CSV

Pattern: lifecycle-timing

Timed outreach based on trigger events.

StepSkillPitch-Packet Behavior
1. Research triggersWeb researchIdentify timing triggers (fiscal year, renewal cycles, etc.)
2. Find contactscompany-contact-finderFind decision-makers at target companies
3. Draft timed sequencesemail-draftingDraft timing-aware email sequences with send-date recommendations

Output: Trigger research document + lead CSV + timing-aware email sequences

Pattern: manual

Strategy requires human execution or tools not yet automated.

StepOutput
1. Generate execution planDetailed step-by-step plan with recommended tools, timelines, and KPIs

Output: Strategy execution plan document (no automated work)

Parallelization

  • Run up to 5 strategy executions concurrently across all companies
  • Track cost per execution against remaining company budget
  • If a single API call would exceed $2, pause and confirm with user

Cost Tracking

Maintain a running cost log:

Cost Tracker:
┌──────────────┬─────────────┬───────┬──────────┬───────────┐
│ Company      │ Strategy    │ Spent │ Budget   │ Remaining │
├──────────────┼─────────────┼───────┼──────────┼───────────┤
│ Acme Corp    │ Phase 1     │ $1.40 │ $5.00    │ $3.60     │
│ Acme Corp    │ Strategy 1  │ $0.70 │          │ $2.90     │
│ Acme Corp    │ Strategy 2  │ $1.10 │          │ $1.80     │
└──────────────┴─────────────┴───────┴──────────┴───────────┘

Output per Strategy

All outputs go to clients/<company>/ in the appropriate subdirectory:

  • Lead CSVs → clients/<company>/leads/
  • Content assets → clients/<company>/content/
  • Email sequences → clients/<company>/campaigns/
  • Strategy plans → clients/<company>/strategies/
  • Signal/research reports → clients/<company>/intelligence/

Phase 4: Packaging

Run client-package-local for each company to produce the final deliverable.

Steps

  1. For each company, run the client-package-local playbook
  2. This produces a dated package directory: clients/<company>/client-package/<YYYY-MM-DD>/ ├── intelligence-package.md ├── growth-strategies.md ├── executed-strategies/ │ ├── strategy-1-signal-outbound/ │ │ ├── signal-report.md │ │ ├── leads.csv │ │ └── email-sequences.md │ ├── strategy-2-competitive-displacement/ │ │ ├── competitor-intel.md │ │ ├── leads.csv │ │ ├── email-sequences.md │ │ └── comparison-page.html │ └──... └── summary.md
  3. Generate a summary.md for each packet that includes:

- Company overview (from intelligence package) - Strategies generated and which were executed - Key metrics: total leads found, content assets created, email sequences drafted - Recommended next steps (what to do if the client signs)

Final Report

After all companies are packaged, present a batch summary:

Batch Complete: 3/3 companies processed

┌──────────────┬────────────┬───────────┬────────┬──────────┬───────────┐
│ Company      │ Strategies │ Executed  │ Leads  │ Content  │ Total Cost│
├──────────────┼────────────┼───────────┼────────┼──────────┼───────────┤
│ Acme Corp    │ 7          │ 4         │ 142    │ 3 assets │ $4.20     │
│ Beta Inc     │ 5          │ 3         │ 89     │ 2 assets │ $3.80     │
│ Gamma Labs   │ 6          │ 4         │ 115    │ 3 assets │ $4.50     │
└──────────────┴────────────┴───────────┴────────┴──────────┴───────────┘

Total cost: $12.50 / $15.00 budget
Packets saved to: clients/<company>/client-package/<date>/

Error Handling

  • Website unreachable: Skip company, flag in report, continue with others
  • Skill failure: Retry once. If still fails, log error, skip that strategy execution, continue with others
  • Budget exceeded: Stop execution for that company, package whatever was completed
  • Rate limiting: Back off and retry with exponential delay (max 3 retries)
  • No strategies tagged executable: Generate all strategies but flag that manual selection will be needed for execution

Configuration Overrides

Users can override defaults at invocation:

Override options:
  budget_per_company: $10      # Default $5
  max_parallel_companies: 5    # Default 3
  max_parallel_strategies: 8   # Default 5
  skip_phases: [4]             # Skip packaging
  strategy_filter: "P0 only"  # Pre-filter strategies
  pitch_packet_mode: false     # Enable live campaign mode (CAREFUL)

WARNING: Setting pitch_packet_mode: false enables live campaign execution — emails will be sent, Apollo credits will be spent. Only do this with explicit user confirmation.

平台分布

Codex

32.17%
按下载量换算23

Claude

32.08%
按下载量换算23

Cursor

19.07%
按下载量换算14

Gemini CLI

8.63%
按下载量换算6

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。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