Healthcare Credentialing Intelligence MCP Server

Healthcare provider credentialing is your first line of defense against fraud, exclusions, and liability. This MCP server delivers structured credentialing intelligence for any physician, nurse practitioner, allied health professional, or facility — screening 8 federal and academic databases in parallel and returning a Composite Credentialing Score (0-100) with a clear verdict and actionable recommendations.

Built for hospital credentialing committees, health insurers, staffing agencies, and compliance officers who need rapid, repeatable provider verification without maintaining multiple database subscriptions. Connect once via the Model Context Protocol and query any of 8 specialized tools directly from Claude, Cursor, or any MCP-compatible AI client.

What data can you access?

Why use Healthcare Credentialing Intelligence?

Manual credentialing is slow and fragmented. Checking OFAC, OIG/LEIE, NPI registries, state licensing boards, Federal Register, PubMed, and corporate registries separately takes a credentialing coordinator 3-6 hours per provider. Errors or omissions create liability: CMS requires hospitals to screen for OIG-excluded providers, and employing an excluded provider can trigger Civil Monetary Penalties of $10,000 per claim.

This MCP server automates the entire multi-database credentialing workflow. An AI agent calls a single tool, waits 30-60 seconds, and receives a structured report with scores across 5 risk dimensions, a composite verdict, and specific recommendations — ready to paste into your credentialing file.

• Scheduling — run monthly re-screening workflows to catch new exclusions and enforcement actions before your next NCQA audit
• API access — trigger credentialing checks from your HRMS, credentialing management system, or custom Python/JavaScript workflow
• Proxy rotation — Apify's built-in infrastructure handles rate limits across all 8 data sources without IP blocks
• Monitoring — get Slack or email alerts when re-screening runs flag new sanctions hits for currently credentialed providers
• Integrations — push results to Zapier, Make, Google Sheets, or webhooks for downstream credentialing workflows

Features

• 8 parallel data source calls — OFAC, OpenSanctions, CFPB, OpenCorporates, Data.gov, Federal Register, ORCID, and PubMed queried simultaneously to minimize latency
• OFAC SDN screening with match confidence — parses numeric match scores (0-100), treats scores ≥80 as high-confidence SDN hits requiring immediate exclusion review
• OIG/LEIE exclusion detection — cross-references Data.gov federal datasets for LEIE, OIG exclusion lists, and provider debarment records using keyword-matched dataset titles
• Multi-jurisdiction sanctions breadth — OpenSanctions dataset array analysis flags entities appearing on 3+ watchlists as multi-jurisdiction exclusion risks
• Federal Register enforcement action parsing — scans regulatory titles and abstracts for 8 exclusion keywords (revoked, debarred, suspended, false claims, kickback) and 9 malpractice keywords (negligence, wrongful death, surgical error, medication error)
• NPDB indicator proxies — Data.gov dataset matching for practitioner adverse action, NPDB, and malpractice reporting datasets
• CFPB complaint severity scoring — distinguishes severe complaints (untimely company response, malpractice keyword in narrative) from general billing complaints with differential weighting
• PubMed recency analysis — counts publications from 2023+ separately, weights recent publications at 5 points vs. 3 points for historical papers
• h-index estimation — approximates h-index from total citation counts using a square-root formula; flags estimated h-index ≥20 as significant research impact
• ORCID identity verification — confirms researcher identity against the Open Researcher and Contributor ID database; absence of ORCID for a provider claiming academic credentials is flagged
• 5-dimensional scoring model — Provider Compliance (25%), Sanctions Exclusion (25%), Malpractice Pattern (20%), License Verification (15%), Publication Activity (15%)
• Composite verdict system — 5 verdict levels: CREDENTIALED, PROVISIONAL, REVIEW_REQUIRED, HIGH_RISK, DO_NOT_CREDENTIAL
• Structured recommendations — auto-generated action text per finding: "Provider appears on exclusion lists — do not credential", "License verification failed — request primary source verification"
• Credential gap analysis — dedicated credential_gap_check tool identifies missing verifications: unconfirmed ORCID, no publications, no corporate records, incomplete license confidence
• Provider comparison — compare_providers tool accepts 2-5 names, runs full analysis on each, and returns ranked results sorted by composite risk score

Use cases for healthcare provider credentialing

Hospital and health system credentialing committees

Credentialing coordinators at acute care hospitals must verify every provider before granting clinical privileges. This MCP automates the initial screening phase: sanctions and exclusion checks, license verification cross-references, and malpractice pattern flags are returned in a structured report that the committee reviews rather than assembles. CMS Conditions of Participation require ongoing monitoring; schedule monthly re-screening for all active medical staff to catch new OIG exclusions automatically.

Health insurer network integrity monitoring

Health plans must continuously monitor their provider networks for newly excluded providers — CMS prohibits payment for services rendered by OIG-excluded individuals. Running provider_compliance_screen on your entire panel quarterly surfaces exclusion risk signals before they become compliance violations. The compare_providers tool supports panel rationalization decisions by scoring candidates side-by-side.

Locum tenens and healthcare staffing agencies

Temporary staffing involves rapid turnaround under time pressure, creating credentialing shortcuts that increase liability. Use facility_credentialing_report for comprehensive new-provider onboarding and credential_gap_check to identify exactly which documents to request before deployment. The compare_providers tool helps rank multiple locum candidates on a single engagement.

Healthcare compliance and legal teams

Compliance officers conducting internal audits, responding to government investigations, or preparing for NCQA/URAC accreditation reviews need documented evidence of credentialing due diligence. This MCP produces structured JSON outputs that log to your audit trail. The Federal Register enforcement action and CFPB complaint data provides early warning of providers who may become subjects of investigation.

Medical staffing due diligence in M&A

When acquiring a physician practice, medical group, or healthcare facility, buyer due diligence must include credentialing status of key providers. Sanctions matches, license revocations, or malpractice patterns for lead physicians materially affect deal value. Run facility_credentialing_report on the top 10-20 providers in the target organization as part of clinical due diligence.

Research institution credentialing and grant administration

Academic medical centers and research institutions credentialing investigators for federally funded research must verify researcher identity, publication record, and absence of debarment. The publication_activity_score tool returns PubMed publication count, ORCID verification status, and an estimated h-index. The credential_gap_check flags missing ORCID verification for researchers claiming academic credentials.

How to screen a healthcare provider for credentialing

1. Connect the MCP server — add the server URL https://healthcare-credentialing-intelligence-mcp.apify.actor/mcp to your Claude Desktop, Cursor, or Windsurf MCP configuration with your Apify API token.
2. Ask your AI assistant — type "Run a full credentialing report on Dr. Sarah Chen at Northside Orthopedics" or use the facility_credentialing_report tool directly with the provider or facility name.
3. Review the composite score and verdict — the response includes a 0-100 risk score, a verdict (CREDENTIALED through DO_NOT_CREDENTIAL), scores across all 5 dimensions, and specific recommendations for any flagged findings.
4. Download or log the structured output — the JSON response can be saved to your credentialing file, pushed to a webhook, or piped into your credentialing management system via the Apify API.

MCP tools

Output example

The facility_credentialing_report tool returns:

{
  "entity": "Dr. Sarah Chen, Northside Orthopedics",
  "compositeScore": 18,
  "verdict": "CREDENTIALED",
  "providerCompliance": {
    "score": 5,
    "sanctionHits": 0,
    "regulatoryFlags": 0,
    "complianceLevel": "COMPLIANT",
    "signals": []
  },
  "sanctionsExclusion": {
    "score": 0,
    "ofacHits": 0,
    "openSanctionsHits": 0,
    "exclusionListMatches": 0,
    "exclusionLevel": "CLEAR",
    "signals": []
  },
  "publicationActivity": {
    "score": 74,
    "peerReviewedCount": 22,
    "orcidVerified": true,
    "hIndexEstimate": 11,
    "activityLevel": "ACTIVE",
    "signals": [
      "22 peer-reviewed publications — strong academic record",
      "8 recent publications (2023+) — actively publishing",
      "ORCID identity verified — academic credentials confirmed"
    ]
  },
  "malpracticePattern": {
    "score": 0,
    "complaintCount": 0,
    "malpracticeIndicators": 0,
    "patternLevel": "CLEAN",
    "signals": []
  },
  "licenseVerification": {
    "score": 68,
    "corporateRecords": 2,
    "regulatoryRecords": 3,
    "confidenceLevel": "VERIFIED",
    "signals": [
      "2 active corporate entities verified"
    ]
  },
  "allSignals": [
    "22 peer-reviewed publications — strong academic record",
    "8 recent publications (2023+) — actively publishing",
    "ORCID identity verified — academic credentials confirmed",
    "2 active corporate entities verified"
  ],
  "recommendations": []
}

A DO_NOT_CREDENTIAL result for a flagged provider:

{
  "entity": "Advanced Pain Solutions LLC",
  "compositeScore": 83,
  "verdict": "DO_NOT_CREDENTIAL",
  "providerCompliance": { "score": 88, "sanctionHits": 3, "regulatoryFlags": 4, "complianceLevel": "EXCLUDED" },
  "sanctionsExclusion": { "score": 76, "ofacHits": 2, "openSanctionsHits": 3, "exclusionListMatches": 2, "exclusionLevel": "EXCLUDED" },
  "malpracticePattern": { "score": 71, "complaintCount": 9, "malpracticeIndicators": 3, "patternLevel": "CRITICAL" },
  "licenseVerification": { "score": 12, "corporateRecords": 0, "regulatoryRecords": 1, "confidenceLevel": "LOW" },
  "allSignals": [
    "3 sanctions/watchlist matches — provider exclusion screening required",
    "OFAC SDN high-confidence match — immediate exclusion review",
    "Entity on 4 watchlists — multi-jurisdiction exclusion risk",
    "2 OIG/LEIE exclusion list matches",
    "5 severe/unresolved complaints — malpractice pattern emerging"
  ],
  "recommendations": [
    "Provider appears on exclusion lists — do not credential",
    "Active sanctions/exclusion match — halt credentialing immediately",
    "Critical malpractice pattern — require enhanced review and peer references"
  ]
}

Output fields

For credential_gap_check, the output includes gaps (string array), gapCount (number), licenseConfidence, publicationActivity, and recommendation (ENHANCED_VERIFICATION_REQUIRED / ADDITIONAL_DOCUMENTATION_NEEDED / CREDENTIALS_SUFFICIENT).

How much does healthcare provider credentialing cost?

This MCP uses pay-per-event pricing — you pay $0.045 per tool call. Platform compute costs are included.

You can set a maximum spending limit per run to control costs. The actor stops when your budget is reached.

Compare this to dedicated credentialing platforms: Verisys charges $3.00-6.00 per NPDB query. Medallion and CredentialStream carry monthly platform fees of $500-2,000+. For AI-assisted workflows that need programmatic credentialing data, this MCP delivers comparable screening data at a fraction of the cost — most users spend under $5/month.

Apify's free tier includes $5 of monthly platform credits, covering your first 100+ tool calls with no payment required.

Healthcare provider credentialing using the API

Python

import requests
import json

APIFY_TOKEN = "YOUR_API_TOKEN"
MCP_URL = "https://healthcare-credentialing-intelligence-mcp.apify.actor/mcp"

headers = {
    "Content-Type": "application/json",
    "Authorization": f"Bearer {APIFY_TOKEN}"
}

# Full credentialing report
payload = {
    "jsonrpc": "2.0",
    "method": "tools/call",
    "params": {
        "name": "facility_credentialing_report",
        "arguments": {"entity": "Dr. Marcus Webb, Westside Cardiology Associates"}
    },
    "id": 1
}

response = requests.post(MCP_URL, headers=headers, json=payload)
result = response.json()
report = json.loads(result["result"]["content"][0]["text"])

print(f"Verdict: {report['verdict']} (score {report['compositeScore']}/100)")
print(f"Compliance: {report['providerCompliance']['complianceLevel']}")
print(f"Exclusion: {report['sanctionsExclusion']['exclusionLevel']}")
print(f"Publications: {report['publicationActivity']['peerReviewedCount']} papers, ORCID verified: {report['publicationActivity']['orcidVerified']}")
print(f"Malpractice: {report['malpracticePattern']['patternLevel']}")
print(f"License confidence: {report['licenseVerification']['confidenceLevel']}")
if report['recommendations']:
    print("Recommendations:")
    for rec in report['recommendations']:
        print(f"  - {rec}")

JavaScript

const APIFY_TOKEN = "YOUR_API_TOKEN";
const MCP_URL = "https://healthcare-credentialing-intelligence-mcp.apify.actor/mcp";

// Compare candidates for a locum tenens engagement
const response = await fetch(MCP_URL, {
    method: "POST",
    headers: {
        "Content-Type": "application/json",
        "Authorization": `Bearer ${APIFY_TOKEN}`
    },
    body: JSON.stringify({
        jsonrpc: "2.0",
        method: "tools/call",
        params: {
            name: "compare_providers",
            arguments: {
                providers: [
                    "Dr. Aisha Okonkwo",
                    "Dr. James Holloway",
                    "Dr. Patricia Reyes"
                ]
            }
        },
        id: 1
    })
});

const result = await response.json();
const comparison = JSON.parse(result.result.content[0].text);

for (const provider of comparison.comparison) {
    console.log(`${provider.provider}: score ${provider.compositeScore} — ${provider.verdict}`);
    console.log(`  Compliance: ${provider.compliance} | Exclusion: ${provider.exclusion} | Malpractice: ${provider.malpractice}`);
}

cURL

# Run a sanctions and exclusion check
curl -X POST "https://healthcare-credentialing-intelligence-mcp.apify.actor/mcp" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -d '{
    "jsonrpc": "2.0",
    "method": "tools/call",
    "params": {
      "name": "sanctions_exclusion_check",
      "arguments": {"entity": "Pinnacle Pain Management Center"}
    },
    "id": 1
  }'

# Run a credential gap check for a specific provider
curl -X POST "https://healthcare-credentialing-intelligence-mcp.apify.actor/mcp" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -d '{
    "jsonrpc": "2.0",
    "method": "tools/call",
    "params": {
      "name": "credential_gap_check",
      "arguments": {"provider": "Dr. Elena Vasquez"}
    },
    "id": 2
  }'

How Healthcare Credentialing Intelligence works

Phase 1: Parallel data collection

When a tool is called, the actor dispatches simultaneous HTTP requests to up to 8 Apify sub-actors using Promise.all. Each sub-actor runs with a 512 MB memory cap and 120-second timeout. The 8 data sources are:

• OFAC Sanctions Search (actor ID: t398ffhJgct2clv9s) — queries the US Treasury SDN list
• OpenSanctions Search (actor ID: 6Zlkz0wjdXewfq3yK) — multi-jurisdiction watchlist including EU, UN, and 100+ national lists
• CFPB Consumer Complaints (actor ID: OuMDkYU3IROUS0AEj) — CFPB complaint database for billing and financial misconduct patterns
• OpenCorporates Search (actor ID: x0a1Q4g0MLc7h15Im) — global corporate registry for entity verification
• Data.gov Dataset Search (actor ID: 1pdOsFEBvCm5RzMfM) — federal open data for NPI registry, OIG/LEIE, and provider enrollment datasets
• Federal Register Search (actor ID: 8bZkbWKlXQrDq0ZgK) — regulatory actions, disciplinary proceedings, and license enforcement
• ORCID Researcher Search (actor ID: Nuq9OYuSRgU3DKFYz) — academic identity verification
• PubMed Research Search (actor ID: AwPvHhEjcgAd6hcvG) — National Library of Medicine publication database

Each tool calls only the subset of sources relevant to its function. publication_activity_score calls only PubMed and ORCID; sanctions_exclusion_check calls OFAC, OpenSanctions, Data.gov, and CFPB.

Phase 2: Dimensional scoring

Five independent scoring functions process the raw returns:

• Provider Compliance — up to 40 pts for OFAC/OpenSanctions hits (15 pts each), 30 pts for Federal Register exclusion keywords, 15 pts for Data.gov OIG/LEIE matches, plus 15-pt compound bonus when sanctions and regulatory flags co-occur.
• Sanctions Exclusion — OFAC SDN max 35 pts (12 pts/hit), OpenSanctions multi-list max 30 pts (10 pts/entity), OIG/LEIE exclusion datasets max 20 pts, CFPB financial misconduct max 15 pts.
• Publication Activity — PubMed papers at 3 pts each + 5-pt recency bonus for 2023+ (max 50), ORCID verification 15 pts, h-index estimate (sqrt of citations) up to 30 pts. Higher = more active; inverted in composite.
• Malpractice Pattern — CFPB complaint count at 2 pts each, severity multiplier of 8 pts for untimely/malpractice-keyword responses (max 35), Federal Register enforcement at 8 pts/action (max 30), NPDB dataset proxies 7 pts each (max 20).
• License Verification — OpenCorporates active entities 8 pts each (max 25), Federal Register license keywords 6 pts each (max 25), Data.gov NPI/DEA hits 6 pts each (max 25), ORCID verification 20 pts. Higher = more verified; inverted in composite.

Phase 3: Composite scoring and verdict

Composite = (Compliance × 0.25) + (Sanctions × 0.25) + ((100 − Publications) × 0.15) + (Malpractice × 0.20) + ((100 − License) × 0.15). License and publication dimensions are positive indicators — a fully verified, actively publishing provider lowers the overall risk score.

Verdict thresholds: 0-19 = CREDENTIALED, 20-39 = PROVISIONAL, 40-59 = REVIEW_REQUIRED, 60-79 = HIGH_RISK, 80-100 = DO_NOT_CREDENTIAL.

Phase 4: Signal aggregation and recommendations

Signals from all five dimensions are consolidated into allSignals. The recommendation engine applies 5 logic rules: EXCLUDED compliance level, EXCLUDED sanctions level, CRITICAL malpractice pattern, UNVERIFIABLE license confidence, and the compound case of inactive publication with incomplete verification. Each match appends a specific action string to the recommendations array.

Tips for best results

1. Use full legal names for highest-confidence screening. "Dr. Sarah Chen" returns fewer false positives than "S. Chen". For organizations, use the full legal entity name as it appears in NPI or state corporate registries.

2. Run credential_gap_check before facility_credentialing_report for new providers. The gap check identifies which documents to request upfront, reducing back-and-forth with providers before you commit to a full report.

3. Use compare_providers when selecting between candidates. The tool runs in a single API call and returns all candidates ranked by composite risk score — more efficient than running individual reports sequentially.

4. Set a budget cap when screening large panels. Use Apify's per-run spending limit when scripting bulk re-screening workflows. At $0.045/call, 500 providers costs $22.50 — set a $25 limit to prevent overruns.

5. Supplement with primary source verification for privileging decisions. This MCP provides risk screening intelligence from public databases. For clinical privilege grants, pair it with primary source verification (state medical board, NPDB, DEA) as required by NCQA/URAC credentialing standards.

6. Schedule quarterly re-screening with Apify's scheduler. OIG exclusions are published monthly; set a quarterly automated run against your credentialed provider list and route DO_NOT_CREDENTIAL or HIGH_RISK results to a compliance alert webhook.

7. Log all credentialing reports. The structured JSON output is audit-ready. Configure a webhook to write each report to your document management system or credentialing database for accreditation documentation.

Combine with other Apify MCP servers

Limitations

• Not a substitute for NPDB primary source query. The NPDB (National Practitioner Data Bank) requires a formal query through HHS for authoritative malpractice payment and adverse action history. This MCP uses Data.gov proxy indicators, not direct NPDB access.
• State medical board data is not directly queried. State-specific license status, disciplinary actions, and board certifications require state board queries. Federal Register and Data.gov data provides partial coverage, not a complete picture of state licensing.
• Publication activity scoring applies to clinician-researchers. Providers in primarily clinical roles (hospitalists, ER physicians) will typically show low publication activity scores. This dimension should be weighted accordingly and not interpreted as a negative credential signal for non-academic roles.
• CFPB complaint data covers financial disputes, not clinical complaints. The CFPB database captures billing, debt collection, and consumer finance complaints. It does not capture state medical board complaints, malpractice suit filings, or clinical adverse events.
• Name matching is string-based. Providers with common names (e.g., "John Smith") may produce false positives from unrelated individuals. Always review the specific signals and data evidence before acting on a high-risk score.
• Data freshness depends on source update frequency. OIG exclusion updates publish monthly; OFAC updates are near real-time; Federal Register data reflects publication delays of 2-4 weeks. Results reflect the state of each source at query time.
• Does not query DEA registration status directly. DEA registration verification currently relies on Federal Register and Data.gov proxy data. For DEA-specific verification, query the DEA Diversion Control Division directly.
• International providers have reduced coverage. OFAC and OpenSanctions provide international coverage, but NPI, OIG/LEIE, and Federal Register data is US-centric. For international medical graduates, coverage is strongest on sanctions/exclusion dimensions.

Integrations

• Zapier — trigger credentialing screens when new providers are added to your HRMS and route results to credentialing coordinators
• Make — build monthly re-screening workflows that pull provider lists from Google Sheets, run screenings, and post flagged results to Slack
• Google Sheets — log credentialing report summaries to a provider compliance tracker with composite scores and verdicts
• Apify API — integrate credentialing checks directly into your credentialing management system or HRMS via HTTP
• Webhooks — push completed reports to your document management system or alert compliance staff when high-risk verdicts are returned
• LangChain / LlamaIndex — use credentialing intelligence as a tool in AI-powered compliance agents or healthcare due diligence workflows

Troubleshooting

• Low scores on known providers. If a legitimate provider returns an UNVERIFIABLE license confidence, this typically means no corporate entity records exist (e.g., providers employed by a hospital rather than operating their own practice entity). Review the licenseVerification.signals field to understand which specific database returned no matches, and supplement with primary source verification.

• High composite score for common names. Providers with common names (e.g., "Michael Chen") may accumulate sanctions and complaint signals from unrelated individuals. Review the allSignals array for entity-specific details. If signals reference different geographic locations or specialties, the matches are likely false positives. Use full names with institutional affiliation (e.g., "Dr. Michael Chen, Northwestern Memorial") to improve specificity.

• Spending limit reached before results. The per-event charge fires before each tool call. If your Apify account balance is insufficient, the tool returns an error object with { "error": true, "message": "Spending limit reached..." }. Add platform credits to your Apify account or raise the per-run budget cap.

• Timeout on facility_credentialing_report. This tool calls all 8 sub-actors simultaneously with a 120-second timeout per actor. Total latency is typically 30-90 seconds. If a sub-actor times out, its result defaults to an empty array and the corresponding dimension scores as zero risk — an optimistic default. For highly time-sensitive queries, use targeted single-dimension tools (sanctions_exclusion_check, provider_compliance_screen) rather than the full report.

• compare_providers returns partial results. If any provider in the comparison array causes an error, that provider's results are excluded from the ranked output. The response will contain fewer entries than the input array. Re-run individual facility_credentialing_report calls for the missing providers.

Responsible use

• This MCP accesses only publicly available government databases and academic registries.
• Credentialing decisions affecting patient care should always include primary source verification in addition to this screening data.
• Comply with applicable employment law when using screening results for hiring or privileging decisions.
• Do not use output to deny care, employment, or privileges based solely on automated screening results without human review.
• For guidance on data use legality, see Apify's guide on web scraping legality.

FAQ

How accurate is healthcare provider credentialing screening against OIG exclusion lists? The sanctions_exclusion_check tool cross-references Data.gov datasets that include OIG/LEIE exclusion data, along with OFAC SDN and OpenSanctions multi-jurisdiction lists. It identifies risk signals and known exclusion dataset matches. For authoritative OIG/LEIE verification, the HHS OIG exclusion database search tool at exclusions.oig.hhs.gov provides the primary source. Use this MCP for programmatic bulk screening and initial risk flagging, then confirm flagged providers against the primary source.

How does healthcare provider credentialing with this MCP compare to Verisys or Medallion? Verisys charges $3.00-6.00 per NPDB query. Medallion and CredentialStream carry monthly SaaS fees of $500-2,000+. This MCP costs $0.045 per call — designed for AI-assisted workflows and compliance automation, not as a full credentialing management platform replacement for large health systems.

Can I use this to credential nurse practitioners, PAs, and allied health professionals? Yes. The tools work with any healthcare provider type by name — physicians, NPs, PAs, pharmacists, dentists, psychologists, and allied health professionals. The same databases (OFAC, OIG/LEIE, Federal Register, NPI) cover all licensed healthcare provider types. Publication activity scoring is most relevant for academic clinicians; set expectations accordingly for primarily clinical providers.

How long does a full facility credentialing report take to run? Typical response time for facility_credentialing_report is 30-90 seconds. The 8 sub-actors run in parallel with a 120-second hard timeout. Network conditions, sub-actor queue times on the Apify platform, and provider name specificity affect total latency. Targeted tools (sanctions_exclusion_check, publication_activity_score) return in 15-45 seconds.

Is it legal to screen healthcare providers using public government data? All 8 data sources are publicly available government databases (OFAC, CFPB, Data.gov, Federal Register) or publicly accessible academic registries (ORCID, PubMed, OpenCorporates). Accessing and processing publicly available data for credentialing and compliance screening is legal in the United States. See Apify's guide on web scraping legality for detailed guidance.

How many providers can I screen in one month on a $5 free credit? At $0.045 per tool call, $5 covers approximately 111 tool calls. Using facility_credentialing_report for full reports, that is 111 complete credentialing screenings. Using targeted tools like sanctions_exclusion_check extends coverage to the same 111 calls but with narrower scope per call.

What does the Composite Credentialing Score measure? The composite score (0-100) is a weighted risk index: Provider Compliance (25%), Sanctions Exclusion (25%), Malpractice Pattern (20%), and inverted License Verification (15%) plus inverted Publication Activity (15%). Publication and license scores are positive indicators — a fully verified, actively publishing provider reduces the composite risk score. A score of 0-19 maps to CREDENTIALED; 20-39 to PROVISIONAL; 40-59 to REVIEW_REQUIRED; 60-79 to HIGH_RISK; 80-100 to DO_NOT_CREDENTIAL.

Can I schedule this MCP to re-screen my provider panel monthly? Yes. Use Apify's built-in scheduler to run a credentialing script against your provider list on any interval. The provider_compliance_screen tool is designed for this use case. Combine it with a webhook to route flagged results to your compliance team. Monthly re-screening for a 200-provider panel costs approximately $9/month.

Does this MCP cover DEA registration status? DEA registration verification relies on Federal Register and Data.gov proxy data. The license_verification tool includes Data.gov queries for NPI and DEA license records. For authoritative DEA registration status, query the DEA Diversion Control Division directly at deadiversion.usdoj.gov. Use this tool to flag providers where DEA-related records appear in regulatory contexts (enforcement actions, revocations in Federal Register).

What happens if the provider name produces no results across all 8 sources? If a provider name returns no data from any source, all dimensional scores default to low risk (no signals = no negative findings). The credential_gap_check tool will flag this explicitly: it returns gaps for missing corporate records, unverified ORCID, and zero publications — prompting you to request primary source documentation. This is the expected behavior for new providers without a digital footprint in federal databases.

Can I connect this MCP to Claude Desktop without writing code? Yes. Add the MCP server URL and your Apify token to your claude_desktop_config.json file (see the connection section below). Once configured, you can ask Claude to run credentialing checks in plain English — no code required.

How is the h-index estimated for publication activity scoring? The h-index estimate uses Math.round(Math.sqrt(totalCitations)) — a simplified proxy based on cumulative citation counts from PubMed results. It is indicative of research impact level and contributes up to 30 points to the publication activity score. A provider with an estimated h-index ≥20 receives a "significant research impact" signal.

How to connect this MCP server

Claude Desktop

Add to your claude_desktop_config.json:

{
  "mcpServers": {
    "healthcare-credentialing-intelligence": {
      "url": "https://healthcare-credentialing-intelligence-mcp.apify.actor/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_APIFY_TOKEN"
      }
    }
  }
}

Cursor, Windsurf, and Cline

Add the server URL https://healthcare-credentialing-intelligence-mcp.apify.actor/mcp with your Apify API token as a Bearer authorization header in your MCP client settings. This MCP is compatible with any client implementing the Model Context Protocol specification.

Programmatic (HTTP)

curl -X POST https://healthcare-credentialing-intelligence-mcp.apify.actor/mcp \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_APIFY_TOKEN" \
  -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"provider_compliance_screen","arguments":{"provider":"Northside Orthopedic Group"}},"id":1}'

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 solutions or enterprise integrations, reach out through the Apify platform.