AegisGate Platform — Enterprise AI Security Gateway

Self-hosted AI security gateway. Six pillars of protection: HTTP API, MCP, A2A, ACP, Response scanning, and Trust Framework. 31 compliance frameworks, 11 SIEM integrations, 4 SOAR integrations. Single 19.1MB Go binary. Deploy in 60 seconds.

🏢 AegisGate Platform v4.4.1 is LIVE — Production hardening, security event categorization, RLS enforcement, gRPC v4 services, 31 compliance frameworks (2,043 total controls, 1,457 automated), 11 SIEM + 4 SOAR integrations, web UI, air-gapped deployment, Guided Setup. Try the Live Demo or Download v4.4.1.

🏢 AegisGate Platform v4.4.1 — canonical facts (source: aegisgate-platform repo)
  • 6 attack surfaces: HTTP API, MCP, A2A, ACP, Response, Trust Framework
  • 31 compliance frameworks (2,043 controls, 1,457 automated): HIPAA, SOC 2, EU AI Act, FedRAMP, ISO 27001, ISO 42001, NIST AI RMF, OWASP LLM Top 10, and more
  • 11 SIEM integrations: Splunk, Elastic, Datadog, Sumo Logic, Chronicle, Loki, Fluentd, Logstash, Cribl, Kafka, Graylog, syslog
  • 4 SOAR integrations: PagerDuty, Jira, ServiceNow, Custom webhook
  • ML threat detection: CharCNN-BiLSTM (1.58M params), 100/100 evasion resistance, 0% FPR, ~5ms
  • 176+ detection patterns, MITRE ATLAS 66 techniques
  • Web UI: 7-page dashboard at /ui/
  • Single 19.1MB Go binary, zero external dependencies, air-gapped deployment
  • 8,000+ tests passing, 83.1% coverage
  • Load tested: 5,000 VUs, 23,578 RPS, 0% errors
  • Apache 2.0 community edition, self-hosted, zero data collection
🏢 AegisGate Platform is open-core (Apache 2.0 community edition + commercial enterprise modules). Free Community tier. Self-hosted. No vendor lock-in. 🚀 Try the Live Demo Download v4.4.1 →

What is AegisGate Platform?

AegisGate Platform is an enterprise AI security gateway. It sits between your organization’s users, agents, and applications — and any AI service they talk to. Every request and response is scanned for security threats, compliance violations, and adversarial attacks.

Who is it for?

  • Security teams who need to enforce AI usage policies across the organization
  • Compliance officers who need audit trails for HIPAA, GDPR, EU AI Act, FedRAMP
  • DevOps teams who need to integrate AI security into CI/CD, SIEM, and SOAR pipelines
  • CISOs who need visibility into what data employees are sending to AI tools

What problem does it solve?

When your team uses AI tools — ChatGPT API, Copilot, MCP-based agents, agent-to-agent communication — they expose your organization to:

  1. Data leakage: Secrets, PII, and IP accidentally sent to external AI providers
  2. Adversarial attacks: Prompt injections that manipulate AI into ignoring safety rules
  3. Compliance violations: Unregulated AI usage that breaks HIPAA, GDPR, EU AI Act
  4. Lack of visibility: No audit trail of what was sent to which AI service

Platform solves all four with a single self-hosted binary.

See Platform in Action

Explore the enterprise dashboard — real-time threat detection, compliance coverage, and SIEM integrations.

Six Pillars of AI Security

Platform v4.4.1 protects every AI interaction across six attack surfaces:

🌐 HTTP API Security

  • 176+ detection patterns
  • MITRE ATLAS 66 techniques
  • PII, secrets, API key detection
  • Bidirectional request/response scanning

🔗 MCP Protocol Protection

  • Session authentication + isolation
  • 8 guardrails active
  • MITRE ATLAS enforcement
  • Tool authorization with risk matrix

🤝 A2A Agent-to-Agent Security

  • mTLS + HMAC-SHA256 integrity
  • Capability enforcement
  • Token-bucket rate limiting
  • Audit logging (RFC 5424)

🛡️ Agent Response Security

  • PII scanner (SSN, cards, emails, phones)
  • Secret detector (Stripe, GitHub, AWS, OpenAI)
  • Hallucination + toxicity detection
  • Response redaction with multiple strategies

🔐 Trust Framework

  • Per-session trust scoring (0–100)
  • Ed25519 signed attestations
  • Cross-pillar correlation
  • Offline-verifiable audit trail

🧠 ML Threat Detection NEW v4.4.0+

  • CharCNN-BiLSTM (1.58M params)
  • 100/100 evasion resistance
  • 0% false positive rate
  • ~5ms inference (pure Go)

Architecture Diagram

flowchart TD subgraph UserLayer["User Layer"] A[Developers] B[AI Agents] C[Applications] end subgraph Platform["AegisGate Platform v4.3.3"] direction TB subgraph Pillar1["🌐 HTTP API Security"] P1[176+ Detection Patterns] P1a[PII & Secrets Detection] P1b[Bidirectional Scanning] end subgraph Pillar2["🔗 MCP Protocol"] P2[Session Auth + Isolation] P2a[8 Guardrails Active] P2b[Tool Authorization] end subgraph Pillar3["🤝 A2A Security"] P3[mTLS + HMAC-SHA256] P3a[Capability Enforcement] P3b[Rate Limiting] end subgraph Pillar4["🛡️ Response Security"] P4[PII Scanner] P4a[Secret Detector] P4b[Response Redaction] end subgraph Pillar5["🔐 Trust Framework"] P5[Trust Scoring 0-100] P5a[Ed25519 Attestations] P5b[Offline Verification] end subgraph Pillar6["🧠 ML Detection"] P6[CharCNN-BiLSTM 1.58M] P6a[100/100 Evasion Resistance] P6b[~5ms Inference] end end subgraph Output["AI Services"] D[OpenAI] E[Anthropic] F[Local LLMs] G[MCP Servers] end A --> Platform B --> Platform C --> Platform Pillar1 --> Output Pillar2 --> Output Pillar3 --> Output Pillar4 --> Output Pillar5 -.->|Attestations| Pillar1 Pillar5 -.->|Attestations| Pillar2 Pillar5 -.->|Attestations| Pillar3 Pillar6 -.->|Threat Scores| Pillar1 Pillar6 -.->|Threat Scores| Pillar4 style Pillar1 fill:#1a1f2e,stroke:#38bdf8,stroke-width:2px style Pillar2 fill:#1a1f2e,stroke:#38bdf8,stroke-width:2px style Pillar3 fill:#1a1f2e,stroke:#38bdf8,stroke-width:2px style Pillar4 fill:#1a1f2e,stroke:#38bdf8,stroke-width:2px style Pillar5 fill:#1a1f2e,stroke:#00ADD8,stroke-width:2px style Pillar6 fill:#1a1f2e,stroke:#22c55e,stroke-width:2px

Detection Flow

sequenceDiagram participant Client as Client App participant Proxy as Platform Proxy participant Scanner as Detection Engine participant Trust as Trust Framework participant SIEM as SIEM/SOAR participant AI as AI Service Client->>Proxy: POST /proxy/ai-request Proxy->>Scanner: Scan Request (PII, Secrets, XSS) alt Threat Detected Scanner-->>Proxy: Block (403 Forbidden) Proxy->>Trust: Generate Attestation Trust-->>Proxy: Signed Attestation Proxy->>SIEM: Log Security Event Proxy-->>Client: 403 Forbidden + Attestation ID else Clean Request Scanner-->>Proxy: Allow Proxy->>AI: Forward Request AI-->>Proxy: AI Response Proxy->>Scanner: Scan Response (PII, Hallucination) alt Response Threat Scanner-->>Proxy: Redact/Block Proxy->>Trust: Generate Attestation Proxy-->>Client: Redacted Response else Clean Response Scanner-->>Proxy: Allow Proxy->>Trust: Generate Attestation Proxy->>SIEM: Log Transaction Proxy-->>Client: AI Response + Attestation end end Note over Proxy,Trust: Every action cryptographically signed Note over Scanner: 176+ patterns + ML model (~5ms)

Compliance Frameworks

Platform includes 31 compliance frameworks (2,043 total controls, 1,457 automated) — the most comprehensive AI compliance coverage available:

FrameworkTierCoverage
OWASP LLM Top 10Community49 patterns
OWASP Web Top 10Community10 categories
MITRE ATLASCommunity66 techniques
NIST AI RMF 1.0Community50 controls
HIPAADeveloperPHI detection (54 controls)
PCI-DSS v4.0DeveloperCard data (152 controls)
SOC 2 Type IIDeveloper64 controls
ISO/IEC 27001:2022Developer116 controls
CCPA/CPRADeveloper26 controls
GDPRDeveloperPII detection (99 controls)
EU AI ActProfessional120 controls, 8 categories
ISO/IEC 42001:2023ProfessionalAI management (38 controls)
NIST CSF 2.0Professional131 controls
CIS Controls v8Professional50 controls
FedRAMP ModerateEnterprise170 controls

...and 16 more frameworks. View all compliance documentation →

SIEM & SOAR Integration

Platform integrates with your existing security operations stack:

SIEM (11 platforms)

PlatformFormatTransport
SplunkHEC JSONHTTPS
Elastic / ELKECS JSONHTTP
DatadogLogs APIHTTPS
Sumo LogicHTTP SourceHTTPS
Chronicle (Google)C2S APIHTTPS
Grafana LokiHTTP PushHTTP
FluentdmsgpackTCP
LogstashBeats / JSONTCP
CriblHTTPHTTPS
KafkaAvro/JSONTCP
GraylogGELFUDP/TCP
Syslog (RFC 5424)StructuredUDP/TCP

SOAR (4 platforms)

PlatformIntegration
PagerDutyIncident trigger with severity mapping
JiraTicket creation with custom fields
ServiceNowIncident + change request
Custom webhookConfigurable payload + headers

For detailed setup, see our SIEM/SOAR Integration Guide.

Web UI

Platform includes a built-in web dashboard at /ui/:

PagePurpose
DashboardReal-time metrics, threat feed, system status
EventsDetection events with filtering and search
ComplianceFramework scan results and reports
SessionsActive AI session monitoring
SettingsConfiguration, API keys, tier management
UsersUser management and RBAC
AuditAudit log viewer with export

Deployment

Guided Setup (30-Second Setup)

# Build the binary
go build -o aegisgate-platform ./cmd/aegisgate-platform/

# Auto-detect your environment and generate a validated config
./aegisgate-platform setup --non-interactive

# Start the platform
./aegisgate-platform --config aegisgate-platform.yaml --embedded-mcp

The setup wizard auto-detects Docker, Kubernetes, systemd, or bare metal; recommends a deploy profile; generates a validated config; and prints next steps. No YAML editing required.

Deploy profiles (5 presets): quickstart, small-team, production, high-security, air-gapped.

# List all profiles
./aegisgate-platform --profile list

# Run with a profile
./aegisgate-platform --profile production --embedded-mcp

docker run -d \
  -p 8080:8080 \
  -p 8081:8081 \
  -p 8082:8082 \
  -p 8083:8083 \
  -p 8443:8443 \
  ghcr.io/aegisgatesecurity/aegisgate-platform:v4.4.0

Bare Metal

# Download the binary
curl -L https://github.com/aegisgatesecurity/aegisgate-platform/releases/latest/download/aegisgate-platform-linux-amd64 -o aegisgate-platform
chmod +x aegisgate-platform
./aegisgate-platform

Kubernetes (Helm)

helm repo add aegisgate https://aegisgatesecurity.github.io/aegisgate-platform
helm install aegisgate aegisgate/aegisgate-platform

Includes HPA autoscaling (1-10 replicas), NetworkPolicy, ServiceMonitor, and rolling updates.

Air-Gapped Deployment

Platform runs fully offline. No external dependencies. No phone-home. No telemetry. The binary, Docker image, and Helm chart can all be transferred to air-gapped environments via offline media.

Pricing Tiers

TierPriceKey Features
CommunityFreeHTTP proxy, MCP, A2A, 176+ patterns, MITRE ATLAS, basic compliance
Developer$79/mo+ mTLS, OIDC/SAML, RBAC, code exec sandbox, advanced ML
Professional$499/mo+ Trust Framework, SIEM/SOAR, gRPC API, K8s/Helm, policy engine
EnterpriseCustom+ Air-gapped deployment, custom compliance modules, dedicated SE, SLA

Full pricing details →

Performance

MetricValue
Sustained RPS23,578
Concurrent VUs tested5,000
Error rate0%
Proxy overhead p992.8ms
PII detection latency94.7 µs
Secret detection latency121.3 µs
ML inference latency~5ms
Binary size19.1 MB
Docker image size19.1 MB
Test coverage83.1%
Tests passing8,000+

Competitive Comparison

CapabilityAegisGateLakera GuardNeMo GuardrailsProtect AI
Self-hosted binary✅❌ (SaaS only)❌ (library)⚠️
MCP protocol protection✅❌❌❌
A2A protocol protection✅❌❌❌
ACP protocol protection✅❌❌❌
Response scanning✅❌⚠️⚠️
Trust attestation✅❌❌❌
ML threat detection✅⚠️❌⚠️
31 compliance frameworks✅❌❌⚠️
11 SIEM integrations✅❌❌⚠️
4 SOAR integrations✅❌❌❌
Air-gapped deployment✅❌❌❌
Open-core (Apache 2.0 community + commercial enterprise)✅❌✅ (Apache)❌

Full competitive analysis →

Deep Dive

For the full technical architecture — six-pillar internals, Trust Framework API, authentication, configuration, and environment variables — see the Technical Architecture page.

Source code → Release notes → Documentation → Pricing → Compliance → Technical architecture →