Skip to main content
Glama
LinkedIn Intelligence & Research MCP Server

LinkedIn Intelligence & Research MCP Server

LinkedIn Intelligence & Research MCP Server

A production-ready Model Context Protocol (MCP) server for LinkedIn Profile & Activity Intelligence, Deep 15-Dimension Post Analysis, Business Signal Detection (GA4, GTM, Meta CAPI), Prospect ICP Matching, and Sales Opportunity Discovery.

Build Status Vitest TypeScript MCP SDK License


📌 Features

  • 🕵️ linkedin_research_profile Orchestrator: High-level command that runs end-to-end profile research, normalizes activity, extracts topics, detects signals, evaluates ICP match, and returns actionable outreach angles.

  • 👤 Static Profile Intelligence (linkedin_analyze_profile): Analyzes headline, about section, role, company size, and positioning.

  • 📝 Deep Post Analyzer (linkedin_analyze_post): 15-dimension post matrix (Topic, Post Type, Hook, Audience, Intent, Tone, CTA, Pain Points, Tools, Sentiment, Commercial Intent).

  • 📡 Business Signal Detector (linkedin_detect_business_signals): Identifies Growth (hiring, expansion), Marketing (funnels, CRM), and Technical Tracking signals (GA4, GTM, Meta CAPI, Attribution).

  • 🎯 ICP Prospect Matcher (linkedin_analyze_prospect): Evaluates profile fit against custom Ideal Customer Profile rules with confidence scoring.

  • 💡 Opportunity Finder (linkedin_find_opportunities): Maps user discussions into technical tracking service opportunities (e.g. GA4 offline conversion tracking).

  • 📅 Activity Timeline (linkedin_build_activity_timeline): Constructs a chronological summary of professional activity over custom date ranges.

  • 📑 MCP Resources (linkedin://): Exposes static/dynamic URIs (linkedin://profiles/{id}, linkedin://profiles/{id}/analysis, linkedin://profiles/{id}/signals).

  • 💬 MCP Prompts: Includes reusable prompt workflows (/profile-analysis, /prospect-analysis, /find-tracking-opportunities).


Related MCP server: signal-prospector

🛠️ Installation & Build

# Install dependencies
npm install

# Run Vitest unit test suite
npm test

# Compile TypeScript to dist/
npm run build

# Start MCP server on stdio transport
npm start

⚙️ MCP Client Configuration

Cursor IDE Configuration

Add to .cursor/mcp.json or Cursor MCP Settings:

{
  "mcpServers": {
    "linkedin-intelligence-mcp": {
      "command": "node",
      "args": ["C:/Users/FLS/.gemini/antigravity/scratch/linkedin-mcp-server/dist/index.js"]
    }
  }
}

Antigravity Configuration

Add to .gemini/antigravity/mcp_config.json:

{
  "mcpServers": {
    "linkedin-intelligence-mcp": {
      "command": "node",
      "args": ["C:/Users/FLS/.gemini/antigravity/scratch/linkedin-mcp-server/dist/index.js"]
    }
  }
}

📄 License

Licensed under the MIT License.

Available Tools

20 tools
linkedin_analyze_postB

Performs a deep 15-dimension analysis on a single LinkedIn post (hook, topic, post type, pain points, tools, sentiment, commercial intent).

Inputs: postId (required), postContent.

linkedin_analyze_profileC

Analyzes static profile headline, about section, experience, skills, and business positioning.

Inputs: profileId (required).

linkedin_analyze_prospectB

High-level prospect analyzer orchestrating profile, company, signals, ICP matching, and lead scoring.

Inputs: profileId (required), icp, objective.

linkedin_analyze_recent_activityC

Analyzes date-range activity, identifies recurring topics, priority shifts, and business signals.

Inputs: profileId (required), to, from, objective.

linkedin_analyze_search_resultsB

Categorizes, filters, and performs preliminary ICP scoring on an imported search result batch.

Inputs: searchId (required), icp.

linkedin_build_activity_timelineC

Constructs a chronological activity timeline summary.

Inputs: profileId (required), to, from.

linkedin_compare_activity_periodsA

Compares activity across two date periods (e.g. July vs. August) to identify strategy shifts, new topics, or disappearing themes.

Inputs: period1To (required), period2To (required), profileId (required), period1From (required), +1 more.

linkedin_create_lead_shortlistC

Generates a curated shortlist of top prospects (e.g. Top 20 PPC agency founders).

Inputs: icp, name, limit.

linkedin_detect_business_signalsA

Scans profile activity to discover growth, marketing, operational, and technical tracking signals (GA4, GTM, CAPI, Attribution).

Inputs: profileId (required), category.

linkedin_extract_topicsB

Extracts top professional topics and mention frequencies from retrieved activity.

Inputs: profileId (required).

linkedin_filter_leadsC

Filters lead batches by minimum score, job title, location, business type, or active tracking opportunities.

Inputs: title, location, minScore, businessType, +1 more.

linkedin_find_opportunitiesC

Identifies specific technical tracking or marketing service opportunities based on observed activity.

Inputs: profileId (required), category.

linkedin_get_recent_activityB

Retrieves normalized activity items over a specified date range.

Inputs: profileId (required), to, from.

linkedin_import_search_resultsA

Ingests JSON/CSV search result objects or profile URL lists into normalized Lead entities.

Inputs: results (required), searchId (required).

linkedin_match_icpC

Evaluates lead fit against custom Ideal Customer Profile rules with confidence scoring.

Inputs: leadId (required), icp.

linkedin_rank_leadsB

Ranks and sorts imported lead batches by overall lead score, tracking opportunities, or intent.

Inputs: limit, sortBy.

linkedin_research_profileA

Runs full end-to-end profile research, normalizes activity, extracts topics, detects signals, evaluates ICP match, and returns actionable outreach angles.

Inputs: profileId (required), to, icp, from, +1 more.

linkedin_score_leadB

Scores a lead using the 100-point Lead Scoring Engine (ICP Fit, Service Relevance, Authority, Signals).

Inputs: leadId (required), icp.

Tool Schema Changelog

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

Loading tool schema changelog…

    TDQS

    B3.2/5.0

    Scored across 20 tools

    Disambiguation3/5

    At least two pairs (research_profile vs analyze_prospect, build_lead_search vs generate_boolean_search) have heavily overlapping scopes, and extract_topics overlaps with analyze_recent_activity. Most other tools are clearly separated by noun and verb, so descriptions resolve most ambiguity, but the overlap is notable.

    Naming Consistency5/5

    All 20 tools follow the linkedin_ prefix plus verb_noun snake_case pattern. Verbs are consistent lowercase and the noun indicates the target resource, making tool names highly predictable.

    Tool Count4/5

    20 tools is on the upper end of typical server size, bordering on heavy. However, the tools form a coherent pipeline spanning activity retrieval, profile/post analysis, signal detection, lead search, scoring, and prospect orchestration, so the count is reasonably justified.

    Completeness4/5

    The server covers the full lead-research workflow: getting and analyzing activity, researching profiles, detecting signals, finding opportunities, building searches, importing/analyzing results, filtering/ranking, and scoring. Minor gaps include standalone company analysis or direct LinkedIn profile lookup, but these are workaroundable.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      D
      maintenance
      Enables research on prospects, pull call transcripts, and enrich leads using Gong, ZoomInfo, Clay, and LinkedIn Sales Navigator through natural language.
      16
      1
      MIT
    • A
      license
      Not graded
      quality
      D
      maintenance
      Enables B2B prospecting from natural language: detect buying signals, score leads against ICP, enrich decision-makers, and draft personalized outreach messages via Claude.
      3
      MIT
    • A
      license
      Not graded
      quality
      B
      maintenance
      Enables AI assistants to securely search your LinkedIn network, contacts, and DMs to find warm leads, intro paths, and hidden opportunities.
      36
      MIT