Security & Trust
Automation you can
prove is safe.
Saltstone agents work inside your most sensitive processes, so security and auditability are part of the platform—not an add-on.
Request a security review
Secure by design
Security is embedded into every agent, integration, and workflow from the first line of code.
Observable by default
Every action is logged, traceable, and reviewable so you can prove what happened and why.
Least privilege
Agents receive only the scoped permissions they need for the task at hand—nothing more.
Security controls
How we protect your data
A layered security model built for agent-driven automation in regulated industries.
Encryption everywhere
Data is encrypted in transit and at rest across every environment we operate, with keys managed through strict access controls.
Least-privilege access
Agents connect through scoped credentials with role-based access, segregation of duties, and just-in-time permissions.
Full decision lineage
Every agent action records its inputs, reasoning, outputs, and human checkpoints for complete auditability.
Data minimization
We process only the fields a workflow requires, with configurable retention windows and safe deletion policies.
Resilient operations
Monitored pipelines, graceful failover, circuit breakers, and human escalation paths protect every workflow.
Compliance alignment
Controls are designed for regulated environments, including healthcare, financial services, and data-protection frameworks.
256-bit
Encryption at rest
TLS 1.3
Data in transit
SOC 2
Aligned controls
99.99%
Target availability
Need our security documentation?
We'll share our control overview and answer every question your review team has.