/ 目錄 / 演練場 / lessie-skill
● 社群 LessieAI ⚡ 即開即用

lessie-skill

作者 LessieAI · LessieAI/lessie-skill

Claude skill for people/company search, contact enrichment, and email outreach via Lessie's CLI — aggregated from public business data.

LessieAI/lessie-skill wraps Lessie's people & company intelligence (CLI + MCP). Data 'aggregated from publicly available sources (business directories, social profiles, corporate websites)'. Commands cover people search by title/company/location, contact enrichment (email + phone), company research (industry/funding/tech stack/hiring), plus Gmail/Outlook email management.

為什麼要用

核心特性

即時演示

實際使用效果

lessie-skill.replay ▶ 就緒
0/0

安裝

選擇你的客戶端

~/Library/Application Support/Claude/claude_desktop_config.json  · Windows: %APPDATA%\Claude\claude_desktop_config.json
{
  "mcpServers": {
    "lessie-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/LessieAI/lessie-skill",
        "~/.claude/skills/lessie-skill"
      ],
      "_inferred": true
    }
  }
}

開啟 Claude Desktop → Settings → Developer → Edit Config。儲存後重啟應用。

~/.cursor/mcp.json · .cursor/mcp.json
{
  "mcpServers": {
    "lessie-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/LessieAI/lessie-skill",
        "~/.claude/skills/lessie-skill"
      ],
      "_inferred": true
    }
  }
}

Cursor 使用與 Claude Desktop 相同的 mcpServers 格式。專案級設定優先於全域。

VS Code → Cline → MCP Servers → Edit
{
  "mcpServers": {
    "lessie-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/LessieAI/lessie-skill",
        "~/.claude/skills/lessie-skill"
      ],
      "_inferred": true
    }
  }
}

點擊 Cline 側欄中的 MCP Servers 圖示,然後選 "Edit Configuration"。

~/.codeium/windsurf/mcp_config.json
{
  "mcpServers": {
    "lessie-skill": {
      "command": "git",
      "args": [
        "clone",
        "https://github.com/LessieAI/lessie-skill",
        "~/.claude/skills/lessie-skill"
      ],
      "_inferred": true
    }
  }
}

格式與 Claude Desktop 相同。重啟 Windsurf 生效。

~/.continue/config.json
{
  "mcpServers": [
    {
      "name": "lessie-skill",
      "command": "git",
      "args": [
        "clone",
        "https://github.com/LessieAI/lessie-skill",
        "~/.claude/skills/lessie-skill"
      ]
    }
  ]
}

Continue 使用伺服器物件陣列,而非映射。

~/.config/zed/settings.json
{
  "context_servers": {
    "lessie-skill": {
      "command": {
        "path": "git",
        "args": [
          "clone",
          "https://github.com/LessieAI/lessie-skill",
          "~/.claude/skills/lessie-skill"
        ]
      }
    }
  }
}

加入 context_servers。Zed 儲存後熱重載。

claude mcp add lessie-skill -- git clone https://github.com/LessieAI/lessie-skill ~/.claude/skills/lessie-skill

一行命令搞定。用 claude mcp list 驗證,claude mcp remove 移除。

使用場景

實戰用法: lessie-skill

Build a B2B prospect list for a sales sequence

👤 Founders, SDRs, agency owners running targeted outreach ⏱ ~45 min intermediate

何時使用: You need 50 qualified contacts for a specific campaign without a Sales Navigator seat.

前置條件
  • Skill cloned + Lessie account — git clone https://github.com/LessieAI/lessie-skill ~/.claude/skills/lessie-skill; sign up at lessie for API key
步驟
  1. Define the ICP
    Use lessie-skill — find Engineering Managers at SaaS companies, 50-500 employees, using Stripe, HQ in EU.✓ 已複製
    → Lessie find-people call with filter JSON, not generic search
  2. Enrich contacts
    Enrich the top 25 with verified email.✓ 已複製
    → Emails plus validation status returned
  3. Draft personalized emails
    Draft individualized openers using each person's recent company news. No spam phrases.✓ 已複製
    → Per-contact personalization, not copy-paste

結果: A usable prospect list with personalized opener drafts.

注意事項
  • Treating 'verified' email as guaranteed deliverable — Warm your sending domain and monitor bounces — verification is probabilistic
搭配使用: github

Source engineering candidates for a specific role

👤 Recruiters and hiring managers on small teams ⏱ ~30 min intermediate

何時使用: You need a shortlist before pinging a recruiter.

步驟
  1. Search
    lessie-skill — find Senior Rust engineers in Berlin or remote EU, with distributed systems background.✓ 已複製
    → Targeted list with signals (seniority, stack evidence)
  2. Qualify
    For the top 10, summarize each in two sentences: strongest fit signal + one reservation.✓ 已複製
    → Honest summaries, not just enrichment fields

結果: A human-readable shortlist ready for outreach.

組合

與其他 MCP 搭配,撬動十倍槓桿

lessie-skill + github

Research GitHub activity of candidates alongside Lessie profile data

For the Rust shortlist, cross-reference each person's GitHub contributions to Rust projects.✓ 已複製

Draft personalized outreach sequences in your brand voice

Using claude-content-writer voice profile, draft 3-email sequences for the top 10 prospects.✓ 已複製

工具

此 MCP 暴露的能力

工具輸入參數何時呼叫成本
find-people title / company / domain / location filters Starting from ICP filters Lessie plan-dependent
enrich-contact person id or profile Shortlist narrowed and ready to contact Per-enrichment credit
find-company company name or domain + optional filters Account research before outreach Plan-dependent
send-email Gmail/Outlook account + to/subject/body Only after you review drafts — don't auto-send at scale Uses your own mailbox

成本與限制

運行它的成本

API 配額
Depends on Lessie plan — the skill itself is just the prompt wrapper
每次呼叫 Token 數
Moderate; responses can be large when listing 50+ contacts
費用
Free — skills are local files. Lessie's data API is paid; consult current plans.
提示
Filter aggressively before enrichment; every enrichment call costs credits.

安全

權限、密鑰、影響範圍

憑證儲存: Lessie API key in env var; Gmail/Outlook via OAuth if you use send-email
資料出站: Queries go to Lessie's API. Email sending goes via your provider.

故障排查

常見錯誤與修復

Skill not invoked — Claude offers generic LinkedIn advice

Say 'use lessie-skill' explicitly.

驗證: ls ~/.claude/skills/lessie-skill/SKILL.md
SKILL.md frontmatter wrong

Re-clone; description drives activation.

驗證: head ~/.claude/skills/lessie-skill/SKILL.md
Wrong trigger keywords — enrichment runs before filtering narrows the list

Explicitly: 'first find-people, then I'll pick who to enrich'.

替代方案

lessie-skill 對比其他方案

替代方案何時用它替代權衡
Apollo.io / ClayYou want a full GTM suite with sequencing toolsMuch heavier platforms; pricier
LinkedIn Sales NavigatorLinkedIn-native prospecting with InMailExpensive, no agent integration

更多

資源

📖 閱讀 GitHub 上的官方 README

🐙 查看未解決的 issue

🔍 瀏覽全部 400+ MCP 伺服器和 Skills