Token导航 LogoToken导航

Agent Skill · 研究检索

tech-stack-teardown(技术栈拆解)

tech-stack-teardown 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。

tech-stack-teardown 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,权限为需要联网;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 72,323。

总安装 188周安装 8GitHub Stars 630下载 66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:tech-stack-teardown(技术栈拆解)
来源仓库:https://github.com/gooseworks-ai/goose-skills
仓库路径:skills/tech-stack-teardown
安装命令:
npx skills add https://github.com/gooseworks-ai/goose-skills --skill tech-stack-teardown
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/gooseworks-ai/goose-skills --skill tech-stack-teardown

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。

用于技术方案的拆解、分析和相关信息检索。适合理解复杂系统的架构组成和依赖关系。可识别关键路径、瓶颈点和改进机会。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。使用前请确认是否依赖架构可视化工具。建议结合领域知识判断拆解深度。

tech-stack-teardown 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/tech-stack-teardown

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/gooseworks-ai/goose-skills --skill tech-stack-…;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

    它需要需要联网,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

tech-stack-teardown 怎么安装?

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

tech-stack-teardown 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。

tech-stack-teardown 需要什么权限?

权限级别是需要联网。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。

tech-stack-teardown 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。

tech-stack-teardown 用的人多吗?

公开数据是累计安装 188 次,近一周安装 8 次,下载 66 次,GitHub 630 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。

tech-stack-teardown 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
tech-stack-teardown

技术栈拆解 · 本页

188630需要联网未明确有提醒

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

Tech Stack Recon

Reverse-engineer a company's sales, marketing, and outbound infrastructure from public signals. No login, no API access to their tools needed — everything is derived from DNS records, website source code, technology profiling, blacklist databases, and public complaints.

What It Detects

CategoryTools Detected
CRMHubSpot, Salesforce (via SPF, website pixels, DNS)
Cold Email ToolsSmartlead, Instantly, Outreach, Salesloft, Lemlist (via SPF, DKIM, TXT records, website source)
People DatabasesApollo, ZoomInfo, Clearbit, 6sense (via website tracker scripts)
Email DeliverySendGrid, Amazon SES, Postmark, Mailgun, Mandrill (via SPF includes, DKIM selectors)
Email MarketingMailchimp, Brevo, ActiveCampaign, Klaviyo (via DKIM selectors)
Ad RetargetingLinkedIn Insight Tag, Facebook Pixel, AdRoll, Reddit Ads, Twitter Ads (via Apify profiler + source)
Website BuilderWebflow, Framer, Next.js, WordPress (via Apify profiler + source)
Chat / SupportIntercom, Drift, Crisp, Zendesk (via website source)
AnalyticsGoogle Analytics, Segment, Mixpanel, Amplitude, PostHog, Heap (via website source)
Outbound DomainsSeparate cold sending domains (via SPF-only Google Workspace + redirect to primary)

How It Works

The skill runs 5 layers of detection, each revealing different signals:

Layer 1: DNS Records (Free, instant)

MX     → Primary email provider (Google Workspace, Microsoft 365, etc.)
SPF    → Every service authorized to send email on their behalf
DKIM   → Cryptographic proof of which tools actually send email
DMARC  → Email authentication policy (how strict they are)
TXT    → Misc verifications (Smartlead tracking domains, tool verifications)
CNAME  → Subdomains pointing to third-party services

This is the highest-signal layer. SPF and DKIM don't lie — if SendGrid is in their SPF, they use SendGrid.

Layer 2: Website Source Inspection (Free, instant)

Fetches the target website and searches HTML for:

  • Tracking pixels (Apollo, REB2B, HubSpot, Facebook, LinkedIn)
  • Script tags loading third-party tools
  • Meta tags and framework signatures
  • Hidden form handlers and API endpoints

Layer 3: Apify Technology Profiler (Pay-per-use, ~$0.005/domain)

Runs justa/technology-profiling-engine actor for deep detection of 7,000+ technologies using 8-tier inspection with confidence scores. Catches tools that don't appear in source code (loaded dynamically, via GTM, etc.).

Layer 4: Blacklist Checks (Free, instant)

Queries 6 major DNS-based blacklists:

  • Spamhaus (zen.spamhaus.org)
  • Barracuda (b.barracudacentral.org)
  • SpamCop (bl.spamcop.net)
  • SORBS (dnsbl.sorbs.net)
  • SURBL (multi.surbl.org)
  • URIBL (black.uribl.com)

Layer 5: Public Complaint Search (Free)

Web searches for spam complaints on Trustpilot, Reddit, SpamCop forums, and general web. Also searches for the company + tool names to find public mentions of their stack.

Cost

ComponentCost
DNS queriesFree
Website source fetchFree
Blacklist checksFree
Web searchesFree
Apify Technology Profiler~$0.005 per domain

Typical costs:

ScenarioDomainsEst. Cost
Single company1~$0.005
Small batch5~$0.025
Large batch20~$0.10

Skip the Apify profiler with --no-apify for free-only analysis (DNS + source + blacklists).

Setup

1. Required

# dig (DNS lookups) — included on macOS/Linux
which dig

# curl (website source fetch) — included on macOS/Linux
which curl

# Python 3 with requests + dotenv
pip3 install requests python-dotenv

2. Optional (for Apify Technology Profiler)

# Get your token at https://console.apify.com/account/integrations
# Add to .env:
APIFY_API_TOKEN=apify_api_YOUR_TOKEN_HERE

Usage

Single Company

python3 scripts/recon.py --domains pump.co

Batch of Companies

python3 scripts/recon.py --domains "dili.ai,pump.co,runautomat.com"

Free-Only Mode (No Apify)

python3 scripts/recon.py --domains pump.co --no-apify

Output to File

python3 scripts/recon.py --domains "dili.ai,pump.co" --output /path/to/report.md

JSON Output

python3 scripts/recon.py --domains pump.co --json

What the Script Does

For each domain:

  1. DNS Scan — Queries MX, SPF, DKIM (18 common selectors), DMARC, TXT records, and 30+ common subdomains (email, tracking, click, bounce, send, smtp, mail, etc.)
  2. Website Source Scan — Fetches the homepage HTML and greps for 40+ known tool signatures (script URLs, pixel IDs, tracking domains)
  3. Apify Technology Profile (optional) — Runs deep 8-tier technology detection for 7,000+ technologies with confidence scores
  4. Blacklist Check — Queries 6 DNS-based blacklists for the domain
  5. Outbound Domain Detection — Checks if common variations of the domain exist (get[name].com, try[name].com, [name]reach.com, etc.) and analyzes their DNS for cold outbound patterns
  6. Report Generation — Produces a structured markdown report with confirmed tools, evidence, email auth assessment, blacklist status, and an overall assessment

Agent Integration

When using this skill as an agent, follow this flow:

  1. User provides one or more company domains
  2. Run recon.py for all domains (confirm Apify cost if > 5 domains)
  3. Present the report — group findings by:

- Confirmed tools (with evidence) - Email authentication (SPF/DKIM/DMARC assessment) - Deliverability (blacklist status + spam complaints) - Notable signals (outbound domains, missing DMARC, SPF gaps)

  1. If batch, include a comparative summary table at the end

Agent Without the Script

The agent can perform all checks manually using built-in tools:

DNS checks — Use Bash tool:

dig +short MX example.com
dig +short TXT example.com
dig +short TXT _dmarc.example.com
dig +short TXT selector._domainkey.example.com
dig +short CNAME subdomain.example.com

Website source scan — Use Bash tool:

curl -sL https://www.example.com | grep -oi 'pattern1\|pattern2\|pattern3' | sort -u

Blacklist checks — Use Bash tool:

dig +short example.com.zen.spamhaus.org A

Apify profiler — Use Bash tool with Python:

# See scripts/recon.py for the full implementation

Spam complaints — Use WebSearch tool:

"example.com" spam OR unsolicited OR "cold email" OR blacklist

DNS Record Cheat Sheet

SPF Includes → Tool Identification

SPF IncludeTool
_spf.google.comGoogle Workspace
spf.protection.outlook.comMicrosoft 365
sendgrid.netSendGrid
amazonses.comAmazon SES
*.hubspotemail.netHubSpot
*.rsgsv.net or servers.mcsv.netMailchimp/Mandrill
spf.mandrillapp.comMandrill (Mailchimp transactional)
mail.zendesk.comZendesk
*.freshdesk.comFreshdesk
spf.mailjet.comMailjet
spf.brevo.comBrevo (Sendinblue)
_spf.salesforce.comSalesforce
mktomail.comMarketo
postmarkapp.comPostmark
mailgun.orgMailgun

DKIM Selectors → Tool Identification

Selector PatternTool
google._domainkeyGoogle Workspace
selector1._domainkey / selector2._domainkeyMicrosoft 365
s1._domainkey / s2._domainkey → *.sendgrid.netSendGrid
k1._domainkey → *.mcsv.net or dkim.mcsv.netMailchimp
k2._domainkey / k3._domainkey → dkim2.mcsv.net / dkim3.mcsv.netMailchimp
mandrill._domainkeyMandrill
pm._domainkeyPostmark
smtp._domainkeyGeneric SMTP
em._domainkeyVarious (check CNAME target)

TXT Records → Tool Identification

TXT PatternTool
open.sleadtrack.comSmartlead (custom tracking domain)
hubspot-developer-verification=*HubSpot
anthropic-domain-verification-*Anthropic (Claude)
MS=*Microsoft 365
google-site-verification=*Google Search Console
slack-domain-verification=*Slack
atlassian-domain-verification=*Atlassian (Jira/Confluence)
docusign=*DocuSign
facebook-domain-verification=*Facebook/Meta
_github-pages-challenge-*GitHub Pages
stripe-verification=*Stripe

Website Source Patterns → Tool Identification

Pattern in HTMLTool
assets.apollo.io/micro/website-trackerApollo.io (visitor tracking)
hs-script or js.hs-scripts.comHubSpot
px.ads.linkedin.comLinkedIn Insight Tag
connect.facebook.net or fbq(Facebook Pixel
snap.licdn.comLinkedIn Insight Tag
cdn.segment.comSegment
cdn.mxpnl.com or mixpanelMixpanel
cdn.amplitude.comAmplitude
app.posthog.com or posthogPostHog
widget.intercom.ioIntercom
js.driftt.comDrift
client.crisp.chatCrisp
static.zdassets.comZendesk
s3-us-west-2.amazonaws.com + reb2bREB2B
clearbit.com/tag.js or revealClearbit Reveal
6sc.co or 6sense6sense
tag.demandbase.comDemandbase
d.adroll.comAdRoll
googletagmanager.comGoogle Tag Manager
gtag('config', 'G-*')Google Analytics 4

Cold Outbound Domain Patterns

A separate domain is being used for cold email if it has:

  • Google Workspace MX (or similar) but no product/marketing email tools in SPF
  • SPF that only includes _spf.google.com (sending from raw mailboxes)
  • A 301/302 redirect to the company's primary domain
  • No website content of its own
  • Domain name follows patterns like: [brand]reach.com, get[brand].com, try[brand].com, meet[brand].com, [brand]hq.com

DMARC Assessment Guide

PolicyMeaningAssessment
p=rejectReject unauthenticated emailStrong — best practice
p=quarantineSend to spam if unauthenticatedGood — enforcing
p=noneMonitor only, don't enforceWeak — anyone can spoof the domain
No DMARC recordNo policy at allMissing — wide open to spoofing

Troubleshooting

"No tools detected"

  • The company may be very early-stage with minimal tooling
  • Some tools (like Apollo used only for prospecting, not sending) leave no DNS trace
  • LinkedIn Sales Navigator, Clay enrichment, and similar tools don't leave public signals
  • Try the Apify profiler if you only ran free-only mode

"SPF has Google only but they use Smartlead/Instantly"

  • This is normal. Smartlead and Instantly typically connect to Google Workspace mailboxes via SMTP and send through Google — so SPF passes via Google's include. The cold email tool itself doesn't need its own SPF entry.
  • Look for Smartlead's open.sleadtrack.com in TXT records or website source as confirmation.

"Apify profiler timed out"

  • Some sites take longer to load. The profiler has a 3-minute timeout.
  • Retry once. If it fails again, rely on DNS + source code analysis.

"artisan.ai resolves but redirects to Afternic"

  • The domain is parked/for sale. Check common alternatives: .co, .com, .io.
  • Wildcard DNS (all subdomains resolve to the same IP) is a sign of a parked domain.

Links

平台分布

Codex

35.06%
按下载量换算23

Claude

30.44%
按下载量换算20

Cursor

16.41%
按下载量换算11

Gemini CLI

8.72%
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

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。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