Skip to main content
Glama
leadpipe-mcp

leadpipe-mcp

by automatiabcn

LeadPipe MCP

AI-powered lead qualification engine for the Model Context Protocol

License: MIT TypeScript MCP

LeadPipe ingests leads from any source, enriches them with company data, scores them 0-100 using configurable AI rules, and exports qualified leads to your CRM — all through the MCP protocol.


Features

  • Lead ingestion from webhooks, forms, APIs, or CSV — single or batch (up to 100)

  • Auto-enrichment with company data: industry, size, country, tech stack (via Hunter.io or domain heuristics)

  • AI scoring engine (0-100) with 6 weighted dimensions + custom rules

  • CRM export to HubSpot, Pipedrive, CSV, or JSON

  • ICP pre-qualification to filter leads on freemail/title/country/tech-stack before spending a single enrichment credit

  • Pipeline analytics with real-time stats, score distribution, conversion rates

  • Configurable scoring weights, high-value titles/industries, custom rules

  • 10 MCP tools + 3 MCP resources covering the full lead lifecycle


Related MCP server: @bingeljell/lead-gen-mcp

Quick Start

Install from MCPize Marketplace

  1. Search for LeadPipe MCP on mcpize.com

  2. Click Install and select your subscription tier

  3. Tools and resources are automatically available in any MCP-compatible client (Cursor, VS Code, etc.)

Build from Source

git clone https://github.com/enzoemir1/leadpipe-mcp.git
cd leadpipe-mcp
npm ci
npm run build

Add to your MCP client config:

{
  "mcpServers": {
    "leadpipe": {
      "command": "node",
      "args": ["path/to/leadpipe-mcp/dist/index.js"]
    }
  }
}

Tools

lead_demo_seed

Seed the pipeline with a realistic demo dataset — 14 leads across 5 archetypes (hot decision-makers, warm mid-level, cold junior/small-co, raw unenriched, and disqualified) — so every downstream tool returns meaningful output without any API keys. Safe to call multiple times; each call appends a fresh batch with new UUIDs.

{}

Returns: counts by status plus sample_lead_ids you can feed into lead_enrich, lead_score, or lead_export.

lead_qualify

Filter leads against your Ideal Customer Profile before spending any enrichment credits. Uses only locally-available signals — email domain, job title, country, industry, company size, tech stack — so nothing is charged to Hunter.io, HubSpot, or Pipedrive.

{
  "criteria": {
    "reject_freemail": true,
    "required_title_keywords": ["vp", "director", "head", "founder"],
    "target_countries": ["US", "CA", "GB"],
    "min_company_size": "11-50",
    "required_tech_stack": ["shopify"]
  },
  "auto_disqualify": true
}

Returns per-lead qualified/rejected decisions with reasons and an estimated credit savings figure.

Pairs well with platform detection tools. If you chain a tool like Detecto (detect_platform) before lead_qualify, the detected tech stack populates company.tech_stack, and required_tech_stack can drop wrong-platform leads before they ever reach enrichment or scoring.

lead_ingest

Add a single lead to the pipeline.

{
  "email": "jane@acme.com",
  "first_name": "Jane",
  "last_name": "Smith",
  "job_title": "VP of Engineering",
  "company_name": "Acme Corp",
  "company_domain": "acme.com",
  "source": "website_form",
  "tags": ["demo-request"]
}

lead_batch_ingest

Add 1-100 leads at once. Duplicates are automatically skipped.

{
  "leads": [
    { "email": "lead1@corp.com", "job_title": "CEO" },
    { "email": "lead2@startup.io", "job_title": "CTO" }
  ]
}

lead_enrich

Enrich a lead with company data using the email domain.

{ "lead_id": "uuid-of-lead" }

Returns: company name, industry, size, country, tech stack, LinkedIn URL.

lead_score

Calculate a qualification score (0-100). Leads scoring 60+ are marked qualified.

{ "lead_id": "uuid-of-lead" }

Returns score + detailed breakdown across all 6 dimensions.

Search and filter leads with pagination.

{
  "query": "acme",
  "status": "qualified",
  "min_score": 60,
  "limit": 20,
  "offset": 0
}

lead_export

Export leads to CRM or file format.

{
  "target": "hubspot",
  "min_score": 60
}

Targets: hubspot, pipedrive, csv, json

Google Sheets export is on the roadmap. Currently returns Sheets-ready formatted data.

pipeline_stats

Get pipeline analytics. No input required.

Returns: total leads, status/source breakdown, average score, score distribution, qualified rate, leads today/week/month.

config_scoring

View or update scoring configuration.

{
  "job_title_weight": 0.30,
  "high_value_titles": ["ceo", "cto", "vp", "founder"],
  "custom_rules": [
    {
      "field": "company_industry",
      "operator": "equals",
      "value": "fintech",
      "points": 15,
      "description": "Bonus for fintech companies"
    }
  ]
}

Resources

Resource

Description

leads://recent

The 50 most recently added leads

leads://pipeline

Pipeline summary with status counts, scores, conversion rates

leads://config

Current scoring engine configuration


Scoring Engine

Leads are scored 0-100 using a weighted average of 6 dimensions:

Dimension

Default Weight

How It Works

Job Title

25%

C-level/Founder: 100, VP/Director: 85, Manager: 65, Senior: 50, Junior: 15

Company Size

20%

Preferred sizes (11-50, 51-200, 201-500): 90, others scaled accordingly

Industry

20%

High-value industries (SaaS, fintech, etc.): 90, others: 40

Engagement

15%

Phone provided, full name, tags, source type (landing page > CSV)

Recency

10%

Today: 100, last week: 75, last month: 35, 3+ months: 5

Custom Rules

10%

User-defined rules with -50 to +50 points each

Formula: score = sum(dimension_score * weight)

Leads with score >= 60 are qualified. Below 60 are disqualified.


CRM Integration

HubSpot

Set the HUBSPOT_API_KEY environment variable with your HubSpot private app access token.

export HUBSPOT_API_KEY="pat-xxx-xxx"

Pipedrive

Set the PIPEDRIVE_API_KEY environment variable.

export PIPEDRIVE_API_KEY="xxx"

CSV / JSON

No configuration needed. Export returns data directly.


Enrichment

LeadPipe extracts the domain from the lead's email and looks up company data:

  1. Hunter.io (if HUNTER_API_KEY is set) — returns organization, industry, country, tech stack

  2. Domain heuristics — maps known domains to company data

  3. Freemail detection — gmail.com, yahoo.com, etc. are flagged (no company enrichment)


Pricing

Tier

Price

Tools

Features

Free

€0

lead_demo_seed, lead_ingest, lead_batch_ingest, lead_search, lead_score, config_scoring

Ingest, manual scoring, ICP pre-qualification

Pro

€19 lifetime

+ lead_qualify, lead_enrich, lead_export, pipeline_stats

AI scoring, Hunter.io enrichment, CRM export, pipeline analytics

One-time €19 lifetime license (3 machines) — no subscription. See Pro License below to buy and activate.


Development

npm run dev        # Hot reload development
npm run build      # Production build
npm test           # Run unit tests
npm run inspect    # Open MCP Inspector

Pro License

LeadPipe ships in Free mode — lead_demo_seed, lead_ingest, lead_batch_ingest, lead_search, lead_score, and config_scoring are open. The following tools require a Pro license:

  • lead_qualify — ICP pre-filter

  • lead_enrich — domain knowledge-base enrichment

  • lead_export — HubSpot / Pipedrive / Google Sheets / CSV / JSON

  • pipeline_stats — portfolio analytics

Buy a Pro License (€19, lifetime, 3 machines): https://automatiabcn.lemonsqueezy.com/buy/360565a3-2577-45e2-93dd-1548a881f456

Or get the Indie MCP Stack Bundle (€69, all 4 servers).

Then activate by setting the env var:

export LEMONSQUEEZY_LICENSE_KEY=YOUR-KEY-HERE

Or in your Claude Desktop / MCP client config:

{
  "mcpServers": {
    "leadpipe-mcp": {
      "command": "npx",
      "args": ["-y", "leadpipe-mcp-server"],
      "env": { "LEMONSQUEEZY_LICENSE_KEY": "YOUR-KEY-HERE" }
    }
  }
}

Validation is cached locally for 24 h, so the server is fully offline-capable after the first run.


License

MIT License. See LICENSE for details.

Built by Automatia BCN.

Available Tools

10 tools
config_scoringScoring ConfigurationA
Idempotent

View or update the global lead scoring configuration used by lead_score. Call with no fields (empty object) to fetch the current config. Pass any subset of fields to patch-update: six dimension weights (each 0–1, should sum to ~1 but not enforced), high_value_titles (string array), high_value_industries (string array), preferred_company_sizes, and custom_rules (array of {name, condition, points}). Changes apply to future lead_score calls only — previously scored leads keep their scores until re-scored.

Inputs: job_title_weight, company_size_weight, industry_weight, engagement_weight, +6 more.

lead_batch_ingestBatch Ingest LeadsA
Idempotent

Add 1 to 100 leads in a single call. Each lead uses the same schema as lead_ingest. Returns {ingested: Lead[], skipped: Array<{email, reason}>} — duplicates are skipped (not failed) so a partial batch still succeeds. Prefer this over repeated lead_ingest calls for bulk imports (CSV/webhook drops).

Inputs: leads (required).

lead_demo_seedSeed Demo LeadsA

Populate the pipeline with a realistic demo dataset: 14 leads across 5 archetypes (hot decision-makers, warm mid-level, cold junior/small-co, raw unenriched, and disqualified). Each lead has appropriate enrichment state, scoring breakdown, and status, so every downstream tool — lead_list, lead_search, lead_score, crm_export, and the pipeline-overview resource — returns meaningful output immediately. Use this to evaluate LeadPipe via MCP Inspector without Hunter, HubSpot, or Pipedrive API keys. Safe to call multiple times; each call appends a fresh batch with new UUIDs. Returns counts by status plus sample_lead_ids you can feed into lead_enrich, lead_score, or crm_export.

Inputs: no named parameters.

lead_enrichEnrich LeadA
Idempotent

Derive and attach company data to an existing lead using the email domain: company name, industry, size, country, website, estimated headcount, and common tech stack. Does not call external APIs — enrichment is driven by the built-in domain knowledge base. Updates the lead in place and returns the enriched record, ready for lead_score. Run this before lead_score for the best qualification accuracy.

Inputs: lead_id (required).

lead_exportExport LeadsA
Idempotent

Push leads to an external destination. target must be one of "hubspot", "pipedrive", "google_sheets", "csv", or "json". For CRM targets (hubspot, pipedrive) the respective API key env var must be set (HUBSPOT_API_KEY, PIPEDRIVE_API_TOKEN) — if missing, the tool returns a dry-run payload instead of erroring. Filter the export via lead_ids (explicit list) or min_score (everything above threshold). Returns {target, count, summary, errors?}.

Inputs: target (required), lead_ids, min_score.

lead_ingestIngest LeadA
Idempotent

Add a single lead to the pipeline. Required: email. Optional: first_name, last_name, job_title, company_name, phone, source ("website"|"linkedin"|"referral"|"event"|"cold_outreach"|"partner"|"other"), tags (string array), custom_fields. Returns the stored lead object with a generated UUID, initial status="new", created_at, and a null score (run lead_score to populate). Throws a duplicate error if the email is already in the pipeline — use lead_search first if you need upsert behaviour.

Inputs: email (required), first_name, last_name, phone, +7 more.

lead_qualifyICP Pre-Qualification (Pre-Enrichment Filter)A
Idempotent

Filter leads against your Ideal Customer Profile BEFORE spending enrichment credits. Uses only locally-available signals (email domain, job_title, country, industry hints, tech_stack) so nothing is charged to Hunter.io, HubSpot, Pipedrive, or any other external service. Set auto_disqualify=true to also update rejected leads to status="disqualified" with the reject reasons stored in custom_fields. If lead_ids is omitted, evaluates every lead currently in status="new". Pairs naturally with upstream platform-detection tools (e.g. Detecto's detect_platform) — run that first to populate company.tech_stack, then run lead_qualify with required_tech_stack=["shopify"] to drop wrong-platform leads before they cost a single API call. Returns qualified/rejected counts, per-lead reasons, and an estimated credit savings figure.

Inputs: criteria (required), lead_ids, auto_disqualify.

lead_scoreScore LeadA
Idempotent

Compute a 6-dimensional qualification score (0-100) for a lead: job_title, company_size, industry, engagement, recency, and custom_rules. Each dimension is weighted via config_scoring; the final score is their weighted average. Updates the lead status to "qualified" (≥60) or "disqualified" (<60) and stores score_breakdown alongside the total. Returns the updated lead with the breakdown. Run lead_enrich first for the most accurate industry/size signals.

Inputs: lead_id (required).

pipeline_statsPipeline StatisticsA
Read-onlyIdempotent

Portfolio-wide pipeline analytics across all leads. Returns {total_leads, leads_today, leads_this_week, leads_this_month, avg_score, qualified_rate (percent), by_status (counts per status), by_source (counts per source), score_distribution}. Takes no input — always aggregates the full dataset. Ideal for dashboards, stand-ups, and conversion-rate tracking.

Inputs: no named parameters.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4.4/5.0

    Scored across 10 tools

    Disambiguation5/5

    Each tool targets a distinct operation: config, ingest (single/batch), demo, enrich, qualify, score, search, export, stats. No two tools overlap in purpose.

    Naming Consistency5/5

    All tools follow a consistent 'lead_<verb>' pattern (e.g., lead_ingest, lead_search), with 'config_scoring' being a minor but still clear exception. The naming is predictable and aids agent understanding.

    Tool Count5/5

    With 10 tools, the set covers the full lead management pipeline without being bloated. Each tool serves a necessary function, making the scope well-balanced.

    Completeness4/5

    The tool set covers the core lifecycle: ingest, enrich, qualify, score, search, export, and config. A notable minor gap is the lack of a direct lead update tool, though the batch ingest and re-scoring can compensate. The demo seed tool is a valuable addition.

    Maintenance

    ActivityStale
    ResponsivenessLimited data

    Related MCP Connectors

    Related MCP Servers