ProductsHow It WorksThe StudioThe Orchestrator
EnterpriseOverviewIntelligenceWho It's ForSecurityIT Governance
Pricing
ResourcesBlogFAQ
Get in touch

Security

Your data stays yours.

Local-first processing, on-device PII masking, and encryption at every layer. Sensitive operations run in a separate secure process that the user interface cannot access.

On-device content detection

A local ML model flags sensitive content before it reaches any AI provider. Three-layer detection: 500+ regex patterns, entropy analysis, and Piiranha ML classification. Runs entirely on the device.

Data masking

PII is detected and masked before any data leaves the device. The transform proxy intercepts all LLM API calls, applies masking, verifies policy compliance, and audit-logs the request before forwarding.

Training opt-in only

Local model training is off by default. Activation requires explicit user consent with EULA acceptance. No silent enrolment. No data sent to third-party AI services for training purposes.

Vendor and model restrictions

Policies define which AI providers and models users can access. Everything outside the allowlist is blocked at the application level. Per-group targeting for different teams.

Architecture

Sensitive operations isolated in a separate secure process

The user interface and the security layer run as independent processes. Data masking, encryption, policy enforcement, and tool routing all happen in the secure process. Even if the application has a vulnerability, protected data stays protected.

Separate secure processA dedicated local process handles everything security-sensitive: data masking, encryption, policy enforcement, tool routing, analytics, and voice processing. The user interface never touches raw sensitive data.
Self-healingThe secure process monitors its own health continuously and recovers from failures without user intervention. If anything goes wrong, it restarts automatically without losing work in progress.
Local tool routingThe MCP gateway multiplexes tool calls between agents and providers. Plugin tools, MCP servers, and terminal operations are all routed through the secure process on the local machine.

Per-agent identity

Each agent authenticates with a token that binds identity, role, repository, and remit. Governance rules define boundaries. Agents check remit before acting and escalate when outside their domain.

Per-session isolation

Sessions are scoped per device. Revoking one device does not affect others. Brute-force protection on all authentication endpoints.

SSO authentication

SAML 2.0, OIDC, Google, and Microsoft. Enterprise controls activate before the workspace loads. Authentication, policy fetch, and legal acceptance must all pass before any AI tools become available.

Policy documentsEd25519 signed, AES-256-GCM encrypted per-device
Key exchangeX25519 per-device key exchange with rotation
TranscriptsAES-256-GCM with Argon2id key derivation
Local stateOS keychain (DPAPI on Windows, Keychain on macOS, libsecret on Linux)
Anti-replayVersion enforcement with nonce-based protection
In transitTLS for all server communication
Pre-sendPII masking on all AI requests
100%AI requests audit-logged
3 layersPII detection (regex + entropy + ML)
Per-deviceRevocation granularity

Compliance

SOC 2 path, GDPR, EU AI Act readiness

Data stored in Zurich, Switzerland. Compliant with Swiss data protection law (FADP) and the EU GDPR. Full audit trails structured for SOC 2 and ISO 27001 reporting.

The EU AI Act requires human oversight, immutable audit trails, decision-point logging, and persistent identity management. Swarmix's structured conversations, proposal voting, plan gates, and agent identity system map directly to these requirements.

Zurich
Data residency
GDPR
Full compliance
ISO 42001
Six clause areas covered
SOC 2
Audit trail structured for reporting

Ready for security review.

We are happy to walk the security team through the architecture, encryption model, and compliance posture.

Book a Call

See how Swarmix works

A walkthrough of unified visibility, governance, and intelligence across every AI vendor and team.

  • Deploy in weeks
  • Works with any existing AI stack
  • SOC 2 compliant, GDPR ready