Skip to main content
HIPAA-Ready • BAA Available • HL7 FHIR R4/R5 • Hash-Chain Audit

API Security for Healthcare
HIPAA-Compliant Protection

The AI Security Layer for healthcare: SHA-256 hash-chain audit for HIPAA §164.312(b), per-tool sandbox for clinical AI agents, 7-year audit retention with WORM Object Lock, and pre-mapped Security Rule controls. All 4 platforms from $299/mo.

HIPAA-Ready (BAA available)
HITECH Act controls implemented
HL7 FHIR R4/R5 Ready
Tamper-evident hash-chain audit
PHI Security Monitor
HIPAA
0
API Calls
0
Blocked
0
PHI Access
Processing
/patients/123Patient Record
read
Audit Log
Waiting for API calls...
HIPAA Compliant
HL7 FHIR Ready
PHI Protection Active
§164.312(b)
hash-chain audit (HIPAA control)
7-year
audit retention (84-month WORM)
BAA
available digitally on paid plans
FHIR R4/R5
resource-level access control
PHI
auto-redaction in security logs
1,700+
threat patterns across 24 categories
OS-level
sandbox for clinical AI agents
99.9%
Enterprise SLA · 1-hr P1

Healthcare API Threats We Prevent

Proactive protection against threats targeting patient data and healthcare systems

PHI Data Exfiltration

Unauthorized access to Protected Health Information (PHI) via EHR/EMR APIs. Behavioral baselines flag abnormal access patterns; anomalies cross +35 risk score and trigger re-auth.

Prevention: Access control validation, rate limiting, anomaly detection on /patients API

FHIR API Abuse

Attackers exploit HL7 FHIR endpoints to scrape records in bulk. Query complexity limits, bundle-size caps, and SMART on FHIR consent enforced before resolution.

Prevention: FHIR query limiting, bundle size restrictions, consent verification

Ransomware API Targeting

Ransomware groups target healthcare APIs to encrypt patient data. Velocity monitoring + adaptive Z-score circuit breaker catches suspicious bulk operations before encryption starts.

Prevention: Velocity monitoring, unusual pattern detection, circuit breakers

Four Platforms for Healthcare

API Security + MCP Security + AI Gateway + Verification Engine — unified under one correlation ID for healthcare workloads

API Security

HIPAA Compliance

Pre-mapped to HIPAA Security Rule §164.312. Cross-framework sync means a SOC 2 control automatically contributes evidence toward HIPAA where they overlap.

  • §164.312(a)(1) - Access control (unique user IDs)
  • §164.312(b) - Audit controls (tamper-evident logs)
  • §164.312(e)(1) - Transmission security (TLS 1.3)

MCP Security

AI Diagnostics

Secure AI agents that assist with clinical decisions. Every tools/call passes 7 sequential checks — permission, MFA, rate limit, rug-pull (SHA-256 drift), threat detect, forwarding, response scan — before touching patient data.

  • Prompt injection detection for clinical AI agents
  • Tool call monitoring for EHR data access
  • HIPAA-compliant audit logs for AI decisions

AI Gateway

Clinical LLMs

Route LLM calls for clinical decision support and medical summarization. PHI scrubbing before sending to LLM providers; adaptive Z-score circuit breaker for failover.

  • Multi-LLM routing (OpenAI, Anthropic, Azure)
  • Automatic failover for critical diagnostics
  • PHI scrubbing before LLM processing

Verification Engine

4 Validation Layers

Validate every clinical AI output before it touches a chart or patient. Real-time enforcement with staged rollout; BLOCK-capable on selected critical paths.

  • PII / PHI masking + forbidden-phrase constraints
  • Source grounding (hallucination detection on clinical claims)
  • FHIR R4/R5 schema validation on AI-generated bundles

Healthcare-Specific Features

Built for healthcare compliance and patient data protection. Every feature pre-mapped to HIPAA, HITECH, and interoperability standards.

Audit-Ready Logs

HIPAA requires audit logs for all PHI access. Every API request appended to a SHA-256 hash chain with monthly partitions, ready for OCR audits and breach investigations.

  • User ID, timestamp, action, correlation ID
  • 7-year (84-month) retention with WORM Object Lock
  • Tamper-evident: full / single / last-N verification
  • Exportable for OCR investigations

Break-the-Glass Access

Allow emergency access to patient data while maintaining audit trails. Critical for ER scenarios where seconds matter.

  • Emergency endpoints bypass rate limits
  • Flagged for post-incident review
  • Automatic security notification
  • Configurable approval workflow

HL7 FHIR Security

Secure FHIR R4/R5 endpoints with resource-level access control. Prevent bulk-export abuse and enforce consent directives at the API layer.

  • Resource-level permissions
  • Search parameter validation
  • Bundle size limiting
  • SMART on FHIR authorization

Patient Consent Management

Enforce patient consent directives at the API layer. Automatically block data sharing for patients who have opted out of specific use cases.

  • Consent directive enforcement
  • Purpose-of-use validation
  • Data segmentation support
  • 42 CFR Part 2 compliance

Breach Detection & Response

Real-time detection of potential PHI breaches with automated containment. Meet HITECH Act breach notification requirements with detailed incident reports.

  • Unusual access pattern detection
  • Auto-block suspicious activity
  • Breach assessment reports
  • 60-day notification tracking

Compliance Reporting

Automated report generation for HIPAA risk assessments, meaningful use attestation, and regulatory audits. Export evidence packages for your compliance team.

  • HIPAA Security Rule mapping
  • Risk assessment automation
  • Evidence package export
  • Control effectiveness scoring

Every Tool Call Passes 7 Sequential Checks

Zero code changes to your EHR or clinical AI stack. Sub-5ms gateway proxy overhead on cached, single-region paths.

1
Permission check
RBAC: does this clinician have access to this patient record?
2
MFA verification
TOTP required for sensitive data (mental health, HIV, substance use, 42 CFR Part 2)
3
Rate limiting
Sliding-window check, Redis-backed with in-memory fallback
4
Rug-pull verification
SHA-256 of tool definition vs. registered hash — block on drift
5
Threat detection
Scan tool arguments for injection patterns + PHI markers
6
Server forwarding
Execute via stdio subprocess, HTTP, or WebSocket transport
7
Response scanning
IndirectInjectionScanner blocks LLM-directed instructions in output
Audit log written
Hash-chain entry: arguments, response, decision, correlation ID
Fail-closed quota state on Redis error • Per-key asyncio lock prevents TOCTOU races • 10 dedicated Prometheus metrics

5 Capabilities You Won't Find Anywhere Else

Not in Anthropic's MCP spec. Not in API gateways. Not in WAFs. Platform-level additions built for clinical-data workloads.

01

OS-Level MCP Sandbox

Subprocess MCP tools execute inside a hardened Linux sandbox. RLIMIT_CPU/AS/NOFILE/NPROC, setsid() process-group isolation, capability dropping, per-tool egress filtering, and shell binaries removed.

modules/mcp/sandbox/executor.py — 934 LOC
02

Tool Definition Hash Registry

SHA-256 hash of every clinical tool definition pinned at tools/list. On every tools/call, the cached definition is re-hashed and compared. Drift raises MCPRugPullDetectedError, blocks execution, publishes a CRITICAL event.

modules/mcp/tool_registry.py • Redis-backed
03

Adaptive Z-Score Circuit Breaker

Statistical, not threshold-based. Z-score > 3.0 against per-hour time-of-day baselines. 4 overlapping sliding windows (1m/5m/15m/1h). Progressive recovery (10→25→50→100%).

gateway/ — 2,208 LOC combined
04

Cross-Pillar Correlation

Every event linked across all four pillars via shared correlation ID. One query: "Show me everything that happened from this encounter — across MCP + API + Gateway + Verification." Architecturally impossible when layers are separate products.

mcp_contexts • parent-child causal chain
05

Hash-Chain Audit System

SHA-256 genesis block, each entry signing the previous. Three verification levels (full / single / last-N). Tamper-evident evidence for HIPAA §164.312(b), SOC 2 CC7.2, and OCR breach investigations.

7 modules • 3,866 LOC combined
+

MCP Correlation Analyzer

Cross-session attack detection: 6-dimension risk score (max 110) across tool sensitivity, data volume, burst, denials, prior detections, and tool diversity. Catches coordinated multi-user attacks and 24h slow-and-low patterns.

MCPCorrelationAnalyzer — alert at score > 50

One Correlation ID. All Four Pillars.

A PHI access event traces forward to the AI tool call it triggered, the EHR API response that returned, and the verification check that caught any drift.

Clinician
Encounter
Prompt
Clinical AI
Tool Call
EHR API
Response
Verification
Recorded in mcp_contexts for parent-child replay • Causal chain reconstruction in one query • Hash-chain entries are tamper-evident for OCR investigations

Healthcare Use Cases

Built for every type of healthcare organization

EHR/EMR Vendors

Secure Epic, Cerner, and custom EHR APIs. Protect patient records from unauthorized access and data breaches.

APIs secured: /patients, /encounters, /medications

Telehealth Platforms

Protect video consultation APIs, prescription endpoints, and patient messaging from abuse and PHI leaks.

APIs secured: /consultations, /prescriptions, /messages

Health Insurance APIs

Secure claims processing, eligibility checks, and benefits verification APIs. Prevent fraud and data manipulation.

APIs secured: /claims, /eligibility, /benefits

Healthcare Platform Integrations

Seamless integration with EHR systems, health information exchanges, identity providers, and compliance tools your organization already uses.

EHR Systems

  • Epic
  • Cerner (Oracle Health)
  • Meditech
  • Athenahealth
  • AllScripts

HIE & Data Exchange

  • CommonWell
  • Carequality
  • eHealth Exchange
  • State HIEs
  • TEFCA QHIN

Identity & SSO

  • Okta
  • Azure AD
  • Ping Identity
  • SMART on FHIR
  • Imprivata

Compliance & GRC

  • Vanta
  • Drata
  • Compliancy Group
  • HITRUST CSF
  • SecurityMetrics

One-Click Epic Integration

Connect G8KEPR to your Epic EHR in minutes. Automatically secure MyChart patient portal APIs, protect FHIR endpoints, and get real-time PHI access alerts without workflow changes.

  • App Orchard marketplace listing
  • Epic FHIR R4 endpoint security
  • MyChart API protection
Integration Status
Epic FHIR APIsSECURED
CommonWell HIECONNECTED
Okta SSOACTIVE

Frequently Asked Questions

Common questions about G8KEPR for healthcare organizations

Audit Evidence, Built In From Day One

Every PHI access event appended to a hash-chain audit log. Pre-built mappings to 11 compliance frameworks with cross-framework sync.

HIPAA-Ready
§164.312 mapped
HITECH
breach response
SOC 2 Type II
observation in progress
GDPR-Ready
Articles 5 / 17 / 32
EU AI Act
controls implemented
FedRAMP
AU-9 evidence

"-Ready" / "controls implemented" reflect capability posture. HIPAA has no third-party certification regime; SOC 2 Type II observation in progress with external audit engagement H2 2026.

Protect Patient Data

Join hospitals and health tech companies using G8KEPR to secure PHI, protect AI diagnostic agents, route clinical LLMs, and satisfy HIPAA Security Rule safeguards with documented evidence.

14-day free trial
HIPAA-Ready
BAA available