Token导航 LogoToken导航

Agent Skill · 研究检索

prospect-enrich(前景丰富)

prospect-enrich 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。

prospect-enrich 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 65,208。

总安装 285周安装 12GitHub Stars 12下载 1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:prospect-enrich(前景丰富)
来源仓库:https://github.com/scientiacapital/skills
仓库路径:skills/prospect-enrich
安装命令:
npx skills add https://github.com/scientiacapital/skills --skill prospect-enrich
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/scientiacapital/skills --skill prospect-enrich

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。

prospect-enrich 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中快速定位候选结果。适用于需要根据关键词或任务场景进行信息搜索与筛选的研究检索场景。通过 npx skills add 命令从指定 GitHub 仓库安装并使用。

安装前建议确认权限范围、维护状态及是否涉及联网或文件操作。具体用法可参考原始 README 和 SKILL.md 文档进一步核验。

来源仓库内路径:skills/prospect-enrich

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/scientiacapital/skills --skill prospect-enrich;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

prospect-enrich 怎么安装?

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

prospect-enrich 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。

prospect-enrich 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。

prospect-enrich 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。

prospect-enrich 用的人多吗?

公开数据是累计安装 285 次,近一周安装 12 次,下载 1 次,GitHub 12 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。

prospect-enrich 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
prospect-enrich

前景丰富 · 本页

28512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

Prospect Enrich Skill

<quick_start> Trigger: Mon-Fri 6:00 AM ET Manual Trigger: "Run prospect enrich" or "Enrich phoneless contacts" Dependencies: Requires HubSpot portal 21530819 access, Apollo credits, Clay credits Output: Branded HTML report + enriched contacts synced to HubSpot with tier tags </quick_start>

<success_criteria>

  • Query HubSpot for all contacts with phone = null (geo: USA/Canada only)
  • Filter out Golden Rule exclusions (customers, channel, device owners, AE-owned, product-page engagers)
  • Identify DEMO REQUEST contacts (first_conversion contains demo/pricing keywords) — enrich FIRST
  • Classify remaining by ATL/BTL tier (see CLAUDE.md ATL/BTL Decision-Maker rules)
  • Batch enrich via Apollo (apollo_people_bulk_match or apollo:enrich-lead)
  • For Apollo misses: send to Clay waterfall (find-and-enrich-contacts-at-company)
  • Sync enriched phone numbers back to HubSpot with tier tags
  • Include company enrichment (domain, revenue, headcount) for ATL-tier contacts
  • Generate branded HTML report with tier sections and Epiphan colors
  • Report: total contacts processed, phone match rate %, ATL/BTL breakdown, credits used </success_criteria>

Stage 1: Pull Phoneless Contacts from HubSpot

MCP Tool: search_crm_objects (HubSpot)

objectType: "contacts"
filterGroups: [{
  filters: [
    { propertyName: "phone", operator: "NOT_HAS_PROPERTY" },
    { propertyName: "hs_lead_status", operator: "IN", values: ["Subscriber", "Qualified Lead", "Inbound Lead", "Marketing Qualified Lead", "Sales Qualified Lead"] }
  ]
}]
properties: [
  "firstname", "lastname", "email", "company", "jobtitle", "phone",
  "first_conversion", "lifecyclestage", "hubspot_owner_id",
  "device_count", "engagement_overview", "is_channel"
]
limit: 100

Output: List of contacts with name, email, company, title, first_conversion; no phone.


Stage 2: Apply Golden Rules Filter

Golden Rules (Hard Exclusions — apply BEFORE enrichment to save credits):

  • EXCLUDE: lifecyclestage = 'customer' (existing customer)
  • EXCLUDE: device_count >= 1 (device owner)
  • EXCLUDE: engagement_overview contains product usage (product-only engager)
  • EXCLUDE: is_channel = true (channel partner)
  • AE-OWNED (90-Day Stale Exception): hubspot_owner_id IN (82625923, 423155215, 190030668)

- If hs_lastmodifieddate > 90 days ago → KEEP but tag as STALE AE LEAD with ATL/BTL tier + deal value. Tim reviews. - Ron Epstein (423155215): 90-day rule applies to ALL leads - Lex Evans (82625923) & Phillip Sandler (190030668): 90-day rule applies to NA contacts only (USA/Canada). Non-NA → still exclude. - If last activity < 90 days → EXCLUDE (AE actively working)

  • EXCLUDE: first_conversion ILIKE '%setup%' — product setup forms only (not sales interest)
  • Keep: Geo filter to USA/Canada only

⚠️ CRITICAL — DO NOT EXCLUDE these first_conversion values:

  • 'Pearl family demo', 'Demo signup', 'demo request'
  • 'General contact us', 'pricing'
  • 'Pearl', 'Connect', 'signup'

These are HIGH-INTENT demo requests — active hand-raisers. Tag them as DEMO REQUEST tier and enrich BEFORE all other tiers.

Output: Filtered list (typically 60-70% of input after Golden Rules)


Stage 3: ATL/BTL Classification Gate (MANDATORY)

Apply BEFORE enrichment to set processing priority order.

Reference: CLAUDE.md § ATL/BTL Decision-Maker Classification v1.0

Tier 1: DEMO REQUEST (Enrich FIRST — Highest Priority)

Any contact whose first_conversion contains:

  • 'Demo signup', 'Pearl family demo', 'demo request'
  • 'General contact us', 'pricing'

These are active hand-raisers — enrich immediately regardless of job title.

Tier 2: ATL (Enrich Second — Budget Authority)

Chief (CIO, CTO, CFO, COO) • Vice President (VP, AVP, SVP, EVP) • President • Provost • Vice Provost • Superintendent • Director (of IT, Technology, Facilities, Academic Technology, Procurement, Materials Mgmt, Medical Education, Court Administration) • Dean • Court Administrator • Clerk of Court (Federal) • City Manager • County Manager • Senior Pastor • Executive Pastor

Tier 3: GRAY (Enrich Third — Flag for Tim's Review)

  • Manager (AV/Facilities/IT) — ATL only if reports to Director+ AND has delegated budget authority >$25K
  • Department Chair — ATL at small institutions; BTL at large universities
  • Director of Educational Technology — depends on reporting line (Provost = ATL; IT VP = maybe BTL)
  • Program Director — ATL only if department-level budget holder; BTL if coordination role

Tier 4: BTL (Enrich Last — Deprioritize)

Technician • Specialist • Coordinator • Support • Administrator (Systems/Network/Database) • Engineer (AV/Network/Systems) • Operator • Instructor/Professor/Faculty • Designer (Learning/Instructional/Graphic) • Assistant • Clerk (non-Court Admin) • Volunteer • Intern • Student • Resident • Help Desk

NEVER ENRICH (Hard Skip — Do NOT Spend Apollo/Clay Credits)

Warehouse Manager • Network Manager • Systems Administrator • AV Technician • Graphic Design Instructor • Program Administrator • Web Designer • Classroom Support • Lab Coordinator • Maintenance • Building Engineer • Multimedia Services Manager • Video Production Specialist • Streaming Crew

Classification Logic:

if first_conversion matches DEMO_KEYWORDS:
  tier = "DEMO REQUEST"  # Enrich first, regardless of title
elif title in NEVER_ENRICH:
  tier = "NEVER"  # Hard skip — do not spend credits
elif title in ATL:
  tier = "ATL"
elif title in GRAY:
  tier = "GRAY"  # Flag for Tim's manual review
else:
  tier = "BTL"

# Processing order: DEMO REQUEST → ATL → GRAY → BTL (skip NEVER)

Stage 4: Batch Enrich via Apollo

Process contacts in tier priority order: DEMO REQUEST → ATL → GRAY → BTL

MCP Tool: apollo_people_bulk_match (or apollo:enrich-lead skill)

details: [
  {
    id: contact.id,
    first_name: contact.firstname,
    last_name: contact.lastname,
    email: contact.email,
    organization_name: contact.company,
    domain: extract domain from email or company website
  }
  ... (up to 10 per batch)
]
reveal_personal_emails: false

Expected Enrichment:

  • phone (direct, mobile, corporate)
  • current_role_min_months_since_start_date (tenure for seniority signal)
  • person_seniorities (c_suite, vp, director, manager, etc.)

Match Rate Target: 65-75% of batch should get phone.


Stage 5: Waterfall Misses to Clay

For Apollo non-matches: Send to Clay find-and-enrich-contacts-at-company

MCP Tool: find-and-enrich-contacts-at-company

companyIdentifier: contact.company (domain preferred)
contactFilters: {
  job_title_keywords: [extract broad role from contact.jobtitle],
  locations: ["United States", "Canada"]
}
dataPoints: {
  contactDataPoints: [{ type: "Email" }],
  companyDataPoints: [{ type: "Latest Funding" }, { type: "Tech Stack" }]
}

Clay Enrichment: Return matching contacts at company with email + company data.


Stage 6: Sync Back to HubSpot

For each enriched contact:

  1. Update HubSpot contact via manage_crm_objects (updateRequest):

- phone (direct or mobile from Apollo/Clay) - hs_analytics_num_page_views (engagement signal) - Tag in notes: [DEMO REQUEST], [ATL], [GRAY], or [BTL]

  1. Enrich company data if not already set:

- MCP Tool: apollo_organizations_enrich - Update HubSpot company: revenue, headcount, founded_date


Stage 7: Generate Branded HTML Report

Output: HTML file saved to workspace as bdr-v3-prospect-enrich-YYYY-MM-DD.html

Styling:

  • Epiphan brand colors: #1a1a2e (dark), #e94560 (accent), #0f3460 (primary)
  • Clean, professional layout
  • Mobile-responsive design

Report Sections:

Summary Metrics Table:

MetricValue
Total phoneless poolX
After Golden Rules filterY (Z% removed)
DEMO REQUEST tierD
ATL tierA
GRAY tierG
BTL tierB
NEVER tier (skipped)N
Phones found (Apollo)PA (PA% of processed)
Phones found (Clay)PC (PC% of non-Apollo)
Total enrichedPA + PC (success %)
Apollo credits usedC

Section 1: DEMO REQUEST Contacts (Green highlight) Sorted first — these are hand-raisers with highest conversion potential.

ContactEmailTitleCompanyPhoneFirst ConversionRevenueAction
...............Demo signup...Priority dial

Section 2: ATL Contacts (Blue highlight) Budget decision-makers — second priority.

Section 3: GRAY Contacts (Yellow highlight) Flagged for Tim's manual review — budget authority uncertain.

Section 4: BTL Contacts (Gray) No budget authority — lowest enrichment priority.

Section 5: NEVER ENRICH Skipped List Contacts that matched NEVER ENRICH titles — credits NOT spent.

Section 6: Key Insights + Recommended Actions

  • Top companies by ATL density
  • Highest-value DEMO REQUESTs
  • Recommended next steps

Stage 8: Emit Outcome Sidecar As the final step, write the following JSON to ~/.claude/skill-analytics/last-outcome-prospect-enrich.json:

{"ts":"[current UTC ISO8601]","skill":"prospect-enrich","version":"1.1.0","variant":"default",
 "status":"[success|partial|error]","runtime_ms":[estimated ms from start],
 "metrics":{"contacts_processed":[Stage 1 total],"phones_found":[Apollo+Clay hits],
 "phone_match_rate_pct":[phones/contacts*100],"atl_count":[ATL tier],"demo_request_count":[DEMO tier],
 "credits_used":[Apollo batch count]},"error":null,"session_id":"[date YYYY-MM-DD]"}

Use status "partial" if some stages failed but results were produced. Use "error" only if no output was generated. Do not include contact names or emails in any field.

Next Steps: Results feed into phone-verification-waterfall-skill for phoneless re-check and prospect-refresh-skill for net-new ICP search.


Skill Metadata

Version: 1.1 Last Updated: 2026-03-19 Author: Tim Kipper Status: Production Integration: HubSpot (portal 21530819) + Apollo + Clay Tier: P1 (Core BDR Automation) Triggers: Scheduled (Mon-Fri 6:00 AM) + Manual ("Run prospect enrich") Changelog:

  • v1.1 (2026-03-19): Added DEMO REQUEST tier (highest priority), AE owner exclusion (IDs 82625923, 423155215), first_conversion filter logic, HTML report output with Epiphan brand colors. Schedule expanded Mon-Fri (was Mon only).
  • v1.0 (2026-03-19): Initial release with ATL/BTL classification, Apollo + Clay waterfall.

平台分布

Codex

35.75%
按下载量换算0

Claude

32.37%
按下载量换算0

Cursor

17.14%
按下载量换算0

Gemini CLI

8.51%
按下载量换算0

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