Corporate Political Exposure MCP Server

Corporate political exposure intelligence for AI agents via the Model Context Protocol. This MCP server queries 11 federal and international data sources in parallel — Senate lobbying, FEC campaign finance, congressional stock trades, FARA foreign agent registrations, SAM.gov contracts, USAspending, OFAC sanctions, OpenSanctions, and OpenCorporates — and applies 5 proprietary scoring models to produce a composite Political Exposure Score (0-100).

This tool is designed for compliance teams, ESG analysts, investment due diligence workflows, and AI agents that need structured political risk intelligence on demand. No subscriptions, no manual research — call a tool, get a scored JSON result.

What data can you access?

MCP tools

Why use this MCP for corporate political risk?

Manual political due diligence requires searching the FEC website, Senate lobbying database, FARA registry, SAM.gov, USAspending, and multiple sanctions lists — each with different interfaces, download formats, and update cycles. For a single company this takes 4-8 hours. For a portfolio of 50 companies, it becomes a dedicated research project.

This MCP server dispatches all queries in parallel, normalizes data across sources, scores every dimension with documented formulas, and returns structured JSON in 30-120 seconds. Your AI agent or application gets an actionable risk score, not a pile of raw government data to interpret.

• Scheduling — run recurring political exposure monitoring on a daily, weekly, or monthly schedule via Apify
• API access — trigger from Python, JavaScript, n8n, or any HTTP client with a token
• Proxy rotation — queries run via Apify infrastructure; no IP blocks or rate limit handling on your side
• Monitoring — configure Slack or email alerts when runs fail or scores exceed thresholds
• Integrations — push results to Zapier, Make, HubSpot, or any webhook endpoint

Features

• 5 proprietary scoring models — Political Dependency (0-100), Influence Network Density (0-100), Legislative Threat/Opportunity (-100 to +100), Foreign Influence Exposure (0-100), and Revolving Door Index (0-100)
• Weighted composite scoring — Foreign influence (25%) + Political dependency (25%) + Influence network (20%) + Revolving door (15%) + Legislative threat (15%) produces a single defensible score
• 11 data sources, all parallel — the influence_network_graph tool fires all 11 actor queries simultaneously and aggregates results; no sequential bottlenecks
• Pay-per-charge billing — every tool call checks Actor.charge() first; your spending cap is enforced at the tool level, not at the run level
• Sanctions escalation logic — any OFAC or OpenSanctions match automatically elevates the Legislative Threat Score by -30 and the Foreign Influence Score by 20-30 points regardless of other signals
• Multi-channel bonus scoring — Influence Network model adds 10 points when 3 or more distinct influence channels (lobbying, campaign finance, congressional trading, foreign agents) are simultaneously active
• Congressional sentiment detection — buy/sell ratio analysis on STOCK Act trades classifies congressional sentiment as BULLISH, BEARISH, or MIXED using a 2:1 ratio threshold with a minimum 3-trade floor
• Bill direction classification — scans bill titles for 10 supportive keywords (support, promote, invest, incentive, authorize) and 6 restrictive keywords (restrict, prohibit, ban, regulate, enforce, penalty) to classify the legislative environment
• Revolving door pattern detection — high lobbying activity combined with government contract presence triggers a "contract-lobbying reinforcement loop" finding; FARA + domestic lobbying combination triggers "international revolving door" detection
• Composite grade labels — EXTREME POLITICAL EXPOSURE (70+), HIGH (50-69), MODERATE (30-49), LOW (0-29) with corresponding actionable recommendations
• Stateless MCP transport — runs in Apify Standby mode via StreamableHTTP at /mcp; each request is fully isolated with no shared session state

Use cases for corporate political exposure intelligence

ESG governance and institutional reporting

ESG governance teams at asset managers and pension funds need standardized political exposure metrics for their portfolio companies. The influence_network_graph tool delivers a scored, structured output that maps directly to the governance pillar of ESG frameworks. Run it quarterly on your holdings list and track score changes as a leading indicator of political risk shifts.

M&A political due diligence

Advisors conducting pre-close due diligence on acquisition targets need to surface hidden political dependencies and influence relationships before deal pricing. Run government_revenue_dependency to quantify federal revenue concentration, foreign_influence_screen to catch FARA and sanctions exposure, and influence_network_graph for the composite view. An elevated political exposure score is a negotiating lever and a disclosure obligation.

Investment research and congressional alpha signals

Quantitative analysts and equity researchers monitor congressional stock trading as a leading signal of legislative intent. The congressional_interest_check tool returns raw buy/sell counts alongside a BULLISH/BEARISH/MIXED sentiment classification for any company. Paired with legislative_threat_assessment, it reveals whether lawmakers are positioning ahead of favorable or restrictive legislation.

Anti-corruption and KYB compliance screening

Compliance officers performing Know Your Business (KYB) screening on counterparties, vendors, and partners need to detect foreign influence connections and PEP-adjacent relationships. The foreign_influence_screen tool hits FARA, OFAC, and OpenSanctions simultaneously, returning a Foreign Influence Exposure Score with individual findings for each source.

Government affairs strategy and monitoring

Government affairs teams at companies with regulatory exposure need to track the legislative environment on an ongoing basis. The legislative_threat_assessment tool classifies the current bill landscape as a STRONG OPPORTUNITY, MILD OPPORTUNITY, NEUTRAL, MILD THREAT, or SEVERE THREAT using real-time congressional bill data and Federal Register activity.

Competitive intelligence and lobbying benchmarking

Strategy teams at corporations and trade associations benchmark their own political influence activity against competitors. Run lobbying_activity_report and campaign_finance_map on multiple companies in a sector and compare lobbying firm relationships, issue coverage, and campaign finance totals side by side.

How to connect this MCP server

Step 1: Get your Apify API token

Sign in at console.apify.com, go to Settings > Integrations, and copy your API token.

Step 2: Add to Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "corporate-political-exposure": {
      "url": "https://corporate-political-exposure-mcp.apify.actor/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_APIFY_TOKEN"
      }
    }
  }
}

Step 3: Configure your AI client

This MCP server works with Claude Desktop, Cursor, Windsurf, Cline, Continue, and any other MCP-compatible client. Use the same endpoint URL and Authorization header pattern.

Step 4: Call a tool

Ask your AI assistant: "Run a political exposure scan on Raytheon Technologies" and the agent will call political_exposure_scan automatically. For a full assessment, ask for an "influence network graph" on any company.

MCP tool reference

All tools accept companyName: string. Tools marked with sector also accept an optional sector: string to broaden legislative bill searches.

Quick example — full assessment:

{ "companyName": "Lockheed Martin", "sector": "defense" }

Output example

The influence_network_graph tool returns structured JSON like this:

{
  "company": "Northrop Grumman",
  "politicalExposureScore": 74,
  "grade": "EXTREME POLITICAL EXPOSURE",
  "models": {
    "politicalDependency": {
      "score": 85,
      "label": "HEAVILY DEPENDENT",
      "findings": [
        "24 SAM.gov contract records — heavy government contract dependency",
        "$2400M in federal awards — extreme government revenue dependency",
        "14 Federal Register references — high regulatory exposure",
        "7 relevant congressional bills — legislative dependency"
      ]
    },
    "influenceNetwork": {
      "score": 72,
      "label": "DEEP NETWORK",
      "findings": [
        "28 lobbying filings — extensive political influence operation",
        "22 FEC contribution records — major political donor",
        "12 congressional stock trades — high lawmaker interest",
        "4 influence channels active — sophisticated political operation"
      ],
      "channels": ["lobbying", "campaign finance", "congressional trading", "foreign agents"]
    },
    "legislativeThreat": {
      "score": 15,
      "direction": "MILD OPPORTUNITY",
      "findings": [
        "Congressional BULLISH: 9 buys vs 3 sells — lawmakers see opportunity",
        "5 supportive vs 1 restrictive bill(s) — favorable legislative environment",
        "28 lobbying filings — industry actively shaping legislation"
      ]
    },
    "foreignInfluence": {
      "score": 35,
      "riskLevel": "MODERATE FOREIGN RISK",
      "findings": [
        "3 FARA registrations — multiple foreign agent ties",
        "6 foreign corporate entities — complex international structure"
      ]
    },
    "revolvingDoor": {
      "score": 65,
      "label": "REVOLVING DOOR ACTIVE",
      "findings": [
        "High lobbying + government contracts — strong revolving door pattern",
        "FARA registrations + domestic lobbying — international revolving door detected",
        "Congressional stock trading + campaign contributions — deep political-corporate ties",
        "Federal awards + active lobbying — contract-lobbying reinforcement loop",
        "5/5 political influence channels active — revolving door ecosystem"
      ]
    }
  },
  "recommendation": "Extreme political exposure. Multiple foreign influence channels, heavy government dependency, and active revolving door patterns. Critical ESG and compliance review required.",
  "dataSources": {
    "lobbying": 28,
    "fecContributions": 22,
    "congressStockTrades": 12,
    "faraRegistrations": 3,
    "congressBills": 8,
    "federalRegister": 14,
    "samContracts": 24,
    "usaSpending": 18,
    "ofacResults": 0,
    "openSanctions": 0,
    "openCorporates": 12
  }
}

Output fields

How the scoring models work

Political Dependency Score (0-100)

Measures reliance on federal government as a revenue source. The model assigns points based on SAM.gov contract record count (up to 30 points for 20+ records), USAspending total obligation amount (up to 25 points for awards over $100M), Federal Register mentions as a regulatory dependency proxy (up to 15 points for 10+ entries), and congressional bill relevance (up to 10 points for 5+ bills). The score is capped at 100. Labels: INDEPENDENT (0-9), LIGHTLY DEPENDENT (10-34), MODERATELY DEPENDENT (35-59), HEAVILY DEPENDENT (60-100).

Influence Network Density Score (0-100)

Maps how many political influence channels are simultaneously active. Points are assigned per channel: lobbying (up to 25 points for 20+ filings), FEC contributions (up to 25 points for 20+ records), congressional stock trading (up to 20 points for 10+ trades), FARA registrations (15 points for any presence), with a 10-point multi-channel bonus when 3 or more channels are active. Labels: MINIMAL NETWORK (0-9), LIGHT NETWORK (10-34), ACTIVE NETWORK (35-59), DEEP NETWORK (60-100).

Legislative Threat/Opportunity Score (-100 to +100)

Classifies the current legislative environment using four signal types. Congressional trade buy/sell ratios using a 2:1 majority threshold contribute ±25 points. Bill title keyword analysis (supportive vs restrictive) contributes ±20 points. Federal Register volume contributes -10 points for heavy regulatory presence. Lobbying intensity contributes +10 points when active lobbying indicates industry engagement. Any OFAC or OpenSanctions match immediately contributes -30 points regardless of other signals.

Foreign Influence Exposure Score (0-100)

Detects foreign government and international entanglement. FARA registrations contribute 10-35 points based on count. OFAC sanctions matches contribute a flat 30 points. OpenSanctions matches contribute 20 points. Foreign corporate entities (non-US jurisdictions in OpenCorporates) contribute up to 15 points for 5+ entities. Labels: MINIMAL (0-4), LOW (5-24), MODERATE (25-49), HIGH FOREIGN RISK (50-100).

Revolving Door Index (0-100)

Detects government-corporate personnel flow patterns through behavioral cross-signals rather than direct officer lookups. High lobbying plus government contracts earns 30 points as a "contract-lobbying reinforcement loop" signal. FARA registrations combined with domestic lobbying earn 20 points as an "international revolving door" signal. Congressional stock trading plus FEC contributions earn up to 20 points for "deep political-corporate ties." Federal awards plus active lobbying earn 15 points. Having 4 or more of the 5 influence channels simultaneously active earns an additional 15 points for "revolving door ecosystem" status.

Composite Political Exposure Score

Weighted average: Foreign Influence (25%) + Political Dependency (25%) + Influence Network (20%) + Revolving Door (15%) + Legislative Threat inverted and normalized (15%). The Legislative Threat score is mapped so that maximum opportunity (+100) contributes 0 to the composite and maximum threat (-100) contributes 100, making the composite score consistently directional — higher always means more exposed.

How much does it cost to assess corporate political exposure?

This MCP server uses pay-per-event pricing with no subscription fees. Individual tools cost $2.00 per call. The full influence_network_graph assessment costs $5.00 because it runs all 11 data sources simultaneously.

You can set a maximum spending limit per run in Apify to cap costs. The server checks this limit before every tool execution and returns a graceful error if the limit is reached. Apify's free tier includes $5 of monthly platform credits, covering one political_exposure_scan or partial credit toward a full assessment at no cost.

Using the API directly

Python

from apify_client import ApifyClient

client = ApifyClient("YOUR_API_TOKEN")

run = client.actor("ryanclinton/corporate-political-exposure-mcp").call(run_input={})

# Or call via HTTP POST to the MCP endpoint after the actor is in Standby mode
import requests

response = requests.post(
    "https://corporate-political-exposure-mcp.apify.actor/mcp",
    headers={
        "Authorization": "Bearer YOUR_API_TOKEN",
        "Content-Type": "application/json",
    },
    json={
        "jsonrpc": "2.0",
        "method": "tools/call",
        "params": {
            "name": "influence_network_graph",
            "arguments": {"companyName": "Palantir Technologies", "sector": "defense"}
        },
        "id": 1,
    },
)
result = response.json()
print(f"Political Exposure Score: {result['result']['content'][0]['text']}")

JavaScript

import fetch from "node-fetch";

const response = await fetch(
    "https://corporate-political-exposure-mcp.apify.actor/mcp",
    {
        method: "POST",
        headers: {
            "Authorization": "Bearer YOUR_API_TOKEN",
            "Content-Type": "application/json",
        },
        body: JSON.stringify({
            jsonrpc: "2.0",
            method: "tools/call",
            params: {
                name: "influence_network_graph",
                arguments: { companyName: "Palantir Technologies", sector: "defense" },
            },
            id: 1,
        }),
    }
);

const data = await response.json();
const parsed = JSON.parse(data.result.content[0].text);
console.log(`Score: ${parsed.politicalExposureScore} — ${parsed.grade}`);
console.log(`Recommendation: ${parsed.recommendation}`);

cURL

# Call the influence_network_graph tool
curl -X POST "https://corporate-political-exposure-mcp.apify.actor/mcp" \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "method": "tools/call",
    "params": {
      "name": "influence_network_graph",
      "arguments": {"companyName": "Lockheed Martin", "sector": "defense"}
    },
    "id": 1
  }'

# Quick scan (faster, $2.00)
curl -X POST "https://corporate-political-exposure-mcp.apify.actor/mcp" \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "jsonrpc": "2.0",
    "method": "tools/call",
    "params": {
      "name": "political_exposure_scan",
      "arguments": {"companyName": "Boeing"}
    },
    "id": 2
  }'

Tips for best results

1. Use sector for legislative tools. The congressional_interest_check and legislative_threat_assessment tools accept an optional sector parameter. For a company like ExxonMobil, passing "sector": "fossil fuels" broadens the congressional bill search to capture industry-level legislative activity that may not name the company directly.

2. Start with political_exposure_scan before committing to a full assessment. The quick scan costs $2.00 and flags sanctions matches, heavy lobbying, and congressional trading. If the scan returns zero flags, the $5.00 full assessment will likely return a low score. Triage first.

3. Run government_revenue_dependency for defense and healthcare companies. These sectors have the highest rates of heavy government contract dependency. The tool returns the dollar total from USAspending, which is often more meaningful than the score alone.

4. Set a spending limit on Apify for automated pipelines. If you are running political exposure scans inside an automated workflow or agent loop, configure a per-run budget cap in the Apify console to prevent runaway costs from agent iteration.

5. For sanctions-risk counterparties, use foreign_influence_screen first. Any OFAC or OpenSanctions hit returns a critical finding immediately. This is the fastest path to a go/no-go decision for sanctions compliance workflows.

6. Combine with periodic scheduling for portfolio monitoring. Configure Apify Schedules to run political_exposure_scan weekly on your top holdings or counterparties. Store results in a dataset and diff the riskFlags array across runs to detect new political risk developments.

7. Pass exact legal entity names for best matching accuracy. Government databases use registered legal names. "Lockheed Martin Corporation" will match more records than "Lockheed." For informal company names, try both the trade name and the legal entity name.

Combine with other Apify MCP servers and actors

Limitations

• Federal data only. This MCP covers federal lobbying (Senate LD filings), federal campaign finance (FEC), and federal contracts (SAM.gov / USAspending). State-level political activity — state lobbying, state PAC contributions, municipal contracts — is not included.
• Historical depth depends on source APIs. FEC and Senate lobbying data extends back multiple election cycles. SAM.gov and USAspending depth varies by source configuration. Real-time accuracy depends on filing and reporting deadlines in each source database.
• Revolving door detection uses behavioral proxies, not personnel databases. The Revolving Door Index detects patterns (high lobbying + government contracts, FARA + domestic lobbying) rather than matching individual officer names against congressional rosters. Direct personnel revolving door confirmation requires a separate HR or executive database.
• Congressional bill keyword scoring is title-based. The legislative threat model scans bill titles for supportive and restrictive keywords. Bills with neutral titles that contain restrictive provisions are not captured without full-text analysis.
• No private equity or subsidiary disaggregation. Queries use the company name as provided. Subsidiaries, holding companies, and recently renamed entities require separate queries with each entity name.
• Response time is 30-120 seconds. All 11 data sources are queried in parallel but each source has its own latency. The influence_network_graph tool in particular should not be used in time-sensitive synchronous workflows.
• Source databases may be incomplete for small companies. Smaller private companies with limited political activity may return low scores not because they have low exposure, but because they have limited public-record presence. Low scores should be interpreted as "limited public record found" for private entities.
• OpenCorporates foreign entity detection is jurisdiction-string based. The foreign corporate structure signal filters out entities where the jurisdiction field contains "us" or "united states." Entities with ambiguous or missing jurisdiction codes may be miscounted.

Integrations

• Zapier — trigger political_exposure_scan from CRM deal stage changes or new counterparty records; push results back to the deal record
• Make — automate scheduled portfolio monitoring workflows that run influence_network_graph on a list of companies and export results to Google Sheets
• Google Sheets — export Political Exposure Scores and dimensional subscores for portfolio-level ESG governance reporting
• Apify API — call tools directly from Python, JavaScript, or any language via the MCP HTTP endpoint or Apify Actor API for ESG platform integrations
• Webhooks — configure alerts when influence_network_graph returns a score above a threshold or when new sanctions flags appear
• LangChain / LlamaIndex — use this MCP server as a tool in LangChain or LlamaIndex agent pipelines for automated political risk research workflows

Troubleshooting

Scores return as 0 or very low for a major company. This most commonly means the company name provided does not match the registered name used in federal databases. Try the full legal entity name. "Microsoft" may miss records filed under "Microsoft Corporation." For conglomerates, try subsidiary names or parent entity names separately.

influence_network_graph times out. The full 11-source query typically completes in 60-120 seconds. If you are hitting a timeout in your MCP client, increase the client timeout setting. Apify actor calls are set to 120 seconds per sub-actor. Very popular company names may return large result sets that extend processing time.

Spending limit reached error. The tool returns "eventChargeLimitReached": true when your configured Apify spending cap is hit. Increase the per-run budget in your Apify actor run settings, or use cheaper tools (political_exposure_scan at $2.00) for initial screening.

FARA results seem low for a foreign-owned company. FARA registration is not required for all foreign-owned businesses — only for entities that act as agents of a foreign government or foreign political party. Foreign-owned commercial companies that do not engage in political activities on behalf of foreign principals are not required to register.

Congressional stock trades return zero for a well-known company. STOCK Act disclosures use individual stock tickers, not company names. The congress-stock-tracker actor queries by company name keyword, so accuracy depends on how congressional members reported their holdings. Large-cap companies with well-known ticker symbols have better coverage than mid-cap or recently public companies.

Responsible use

• All data sources used by this MCP are US government public databases (FEC, Senate Lobbying Disclosure, FARA, SAM.gov, USAspending, Federal Register, Congress.gov) and established international compliance databases (OFAC, OpenSanctions, OpenCorporates).
• OFAC and OpenSanctions data should be used for informational screening only. Compliance decisions should be confirmed against the official source databases before action is taken.
• This tool does not constitute legal advice. Political exposure assessments should be reviewed by qualified compliance counsel before use in regulatory filings, investment decisions, or legal proceedings.
• Comply with GDPR and applicable data protection regulations when storing or processing data about individuals identified through FARA registrations, FEC contributions, or congressional stock trade disclosures.
• For guidance on web scraping and data use legality, see Apify's guide.

FAQ

How does corporate political exposure scoring work across the 5 models? The server runs up to 11 parallel data source queries per assessment. Each of the 5 scoring models processes a subset of those results using documented point thresholds (see the "How the scoring models work" section above). The composite score is a weighted average: foreign influence and political dependency each contribute 25%, influence network contributes 20%, revolving door 15%, and normalized legislative threat 15%.

How accurate is the Political Exposure Score for private companies? Accuracy depends on federal record presence. Large private companies with government contracts and PAC activity have strong federal footprints and score accurately. Small private companies with no federal contracts, no lobbying filings, and no FEC activity will return low scores that reflect limited public record rather than confirmed low exposure.

How long does a full influence_network_graph assessment take? Typically 30-90 seconds. All 11 data source queries run in parallel using Promise.all. The bottleneck is the slowest individual source. On Apify infrastructure each sub-actor is allocated 256MB memory and a 120-second timeout.

Can I run corporate political exposure assessments on a schedule? Yes. Configure Apify Schedules to call the tool on daily, weekly, or monthly intervals. Store results in an Apify dataset and use dataset comparison or webhooks to detect score changes over time.

Does this MCP detect corporate political exposure for non-US companies? The data sources are primarily US federal databases, so coverage is strongest for US-domiciled entities and foreign companies with US operations, lobbying activity, or FARA registrations. The foreign_influence_screen tool is most useful for identifying US entities with foreign entanglements rather than for profiling foreign companies directly.

How is this different from the Counterparty Due Diligence MCP? This MCP is specialized for political risk dimensions: lobbying, campaign finance, congressional activity, government revenue dependency, and foreign influence. The Counterparty Due Diligence MCP covers broader KYB screening including corporate registry verification, SEC filings, and digital footprint analysis. For full counterparty screening, run both.

What does the Revolving Door Index actually measure? The Revolving Door Index uses behavioral cross-signals rather than individual personnel matching. High lobbying activity combined with government contract presence signals a contract-lobbying reinforcement loop — a common revolving door pattern. FARA registrations plus domestic lobbying signals an international revolving door. Having 4 or more of the 5 influence channels simultaneously active signals a revolving door ecosystem. It is a structural pattern detector, not an individual name matcher.

Is it legal to use this data for compliance screening? All underlying data sources are public US government databases. Using them for compliance screening is legal and standard practice. OFAC and OpenSanctions data are specifically intended for sanctions compliance. FEC, Senate lobbying, and FARA data are published by the US government for public transparency. For GDPR-covered entities, review data minimization obligations before storing individual-level records. See Apify's legality guide.

Can I use the MCP endpoint from an AI agent framework like LangChain? Yes. The /mcp endpoint implements the standard Model Context Protocol over Streamable HTTP. Any MCP-compatible client library can connect to it. The endpoint URL is https://corporate-political-exposure-mcp.apify.actor/mcp and requires a Bearer token in the Authorization header.

What happens if a data source returns no results? Each sub-actor call is wrapped in a try-catch that returns an empty array on failure. The scoring models handle zero-length arrays gracefully — a source returning no results contributes 0 points to the relevant model, and findings will note "No [X] detected." The composite score reflects available data, not a failure state.

Can I request additional data sources or scoring models? Yes. Submit feature requests via the Issues tab on this actor's page. Priority candidates include state-level lobbying databases, EU transparency registers, and UK Electoral Commission donation records.

Help us improve

If you encounter issues, you can help us debug faster by enabling run sharing in your Apify account:

1. Go to Account Settings > Privacy
2. Enable Share runs with public Actor creators

This lets us see your run details when something goes wrong, so we can fix issues faster. Your data is only visible to the actor developer, not publicly.

Support

Found a bug or have a feature request? Open an issue in the Issues tab on this actor's page. For custom scoring models, enterprise integrations, or portfolio-level monitoring solutions, reach out through the Apify platform.