Leads Toolkit
БесплатноНе проверенMCP servers for local-business prospecting: Google Places search, automated website tech/SEO gap scoring, and lead tracking — orchestrated by Claude across inde
Описание
MCP servers for local-business prospecting: Google Places search, automated website tech/SEO gap scoring, and lead tracking — orchestrated by Claude across independent, single-purpose servers.
README
A suite of MCP servers for finding, qualifying, and tracking small-business leads. Built for solo freelance prospecting at nickfurr.com — general lead prospecting for local service businesses and SMBs.
Servers
| Server | Status | Purpose |
|---|---|---|
| leads-search | shipped | Search Google Places for businesses by keyword + location |
| leads-enrich | shipped | Analyze a lead's website for tech signals, SEO gaps, and opportunity score |
| leads-store | scaffold | Persist leads to a per-project JSON file: save, query, tag, mark contacted, export |
Each server is an independent MCP process. Claude orchestrates them — no server calls another. The data plane between them is a JSON file on disk that leads-store owns.
Typical Workflow
search_leads("coffee shops", "Austin, TX", saveAs: "austin-coffee")
list_leads({ tags: ["austin-coffee"] })
→ Claude iterates: deep_enrich(url, "websites", { placeId }) on each
list_leads({ tags: ["austin-coffee"], minScore: 60 })
→ top tier; find_contacts(url, { save: { placeId } }) on each
export_leads({ tags: ["austin-coffee"], hasEmail: true }, "csv", "./outreach.csv")
Setup
Prerequisites
- Node.js 18+ (developed on 22.x).
node --versionto check. - Claude Desktop installed.
- Google Cloud project with both Places API and Geocoding API enabled, plus an API key — required by
leads-search.leads-enrichandleads-storeneed no external keys.
1. Install dependencies (per server)
node_modules/ is gitignored, so a fresh clone needs an install in each server folder:
cd leads-search && npm install
cd ../leads-enrich && npm install
cd ../leads-store && npm install
2. Wire up Claude Desktop
Edit %APPDATA%\Claude\claude_desktop_config.json (Windows) or ~/Library/Application Support/Claude/claude_desktop_config.json (macOS). A working template lives at leads-enrich/claude_desktop_config.example.json — copy its mcpServers block into your config and replace:
/ABSOLUTE/PATH/TO/leads-toolkit/...with the actual absolute path on this machine (Claude Desktop does not resolve relative paths).your-google-api-key-herewith your real Google Places key.
Optional env vars:
LEADS_STORE_PATHonleads-store— overrides the default.leads.jsonresolution (cwd → ancestor dirs →%USERPROFILE%\.leads\default.json).
3. Restart Claude Desktop
You should see search_leads, get_lead_details, and the leads-enrich tools in the MCP menu. (leads-store is scaffold-only — its tools are not yet implemented.)
Data note
.leads.json files are gitignored — they don't move with the repo. If you want existing lead data on a new machine, copy the relevant .leads.json files manually.
Design
The full design for the upcoming leads-store server, the multi-page contact scraping in leads-enrich, and the configurable scoring profiles is in docs/specs/2026-05-07-leads-pipeline-design.md.
Установка Leads Toolkit
У этого сервера нет опубликованного пакета — он собирается из исходников. Открой репозиторий и следуй инструкции в README.
▸ github.com/nick-furr/leads-toolkitFAQ
Leads Toolkit MCP бесплатный?
Да, Leads Toolkit MCP бесплатный — установка в пару кликов через Unyly без оплаты.
Нужен ли API-ключ для Leads Toolkit?
Нет, Leads Toolkit работает без API-ключей и переменных окружения.
Leads Toolkit — hosted или self-hosted?
Self-hosted: сервер запускается локально на твоей машине командой из раздела установки.
Как установить Leads Toolkit в Claude Desktop, Claude Code или Cursor?
Открой Leads Toolkit на unyly.org, выбери вкладку своего клиента (Claude Desktop, Claude Code, Cursor) и нажми Install — конфиг сгенерируется автоматически, без правки JSON.
Похожие MCP
Fetch
Web content fetching and conversion for efficient LLM usage.
AWS KB Retrieval
Retrieval from AWS Knowledge Base using Bedrock Agent Runtime.
автор: modelcontextprotocolSpring AI MCP Server
Provides auto-configuration for setting up an MCP server in Spring Boot applications.
llm-analysis-assistant
A very streamlined mcp client that supports calling and monitoring stdio/sse/streamableHttp, and can also view request responses through the /logs page. It also
автор: xuzexin-hzCompare Leads Toolkit with
Не уверен что выбрать?
Найди свой стек за 60 секунд
Автор?
Embed-бейдж для README
Похожее
Все в категории ai
