Trust Center

Security & Trust at Saltbox

Your Salesforce delivery data deserves the same level of protection you give your production org. Security isn't an afterthought at Saltbox — it's built into every layer of the platform.

SOC 2 Type IIIn progress
GDPR Compliant

Data Protection

Defense in depth, by design

Encryption everywhere

All data is encrypted in transit using TLS and at rest using industry-standard AES-256 encryption. Integration credentials — Salesforce tokens, API keys, OAuth secrets — are encrypted with rotating keys and are only decrypted at the moment they're needed.

They're never logged, never cached, and never exposed to client-side code.

Tenant isolation

Saltbox is multi-tenant by design. Every query, every API call, every background job is scoped to your organization. Your data is architecturally separated from every other customer's data — not just by application logic, but at the data layer.

Within your organization, access is further scoped by project membership. Users only see the projects they've been added to. Identity is always derived server-side from your authenticated session, never from request parameters.

Infrastructure

Enterprise-grade cloud hosting

Saltbox runs on trusted cloud infrastructure including Vercel, Heroku, PostgreSQL, and Google Cloud — providers that maintain their own SOC 2, ISO 27001, and GDPR certifications.

  • HTTPS only. All traffic is encrypted via HTTPS — no unencrypted connections accepted.
  • Hardened headers. Industry-standard security headers protect against clickjacking, MIME sniffing, and other common web vulnerabilities.
  • Global CDN. Distribution with automatic failover for high availability.
  • Automated backups. Database backups with point-in-time recovery.

Authentication & Access

Secure by default

SSO & API keys

Saltbox supports Single Sign-On via Google OAuth for secure, passwordless authentication. API keys are available for service-to-service integrations — each scoped to specific permissions with built-in expiration and revocation.

Role-based access

A three-tier role model — Member, Admin, and Owner — ensures users have exactly the access they need and nothing more. Project-level membership adds a second layer, so sensitive work stays visible only to the people involved.

Rate limiting

All API endpoints are protected by adaptive rate limiting to guard against abuse while keeping legitimate usage uninterrupted.

AI Security

Your data, your control

Saltbox uses AI to accelerate Salesforce project delivery. Here's how we keep your data safe throughout.

No model training on your data.

Your conversations, project data, and Salesforce metadata are never used to train AI models. Our commercial agreements with AI providers explicitly prohibit this.

Isolated context per session.

Each AI conversation operates in a sandboxed environment scoped to your project. The AI cannot access data from other projects or organizations.

Human-in-the-loop.

Sensitive operations like Salesforce deployments and record creation require explicit user approval. The AI proposes — you decide.

Full transparency.

Every AI action is visible in the chat interface. There are no hidden operations or background calls you can't see.

Comprehensive logging.

All AI interactions are monitored for quality, performance, and security through our observability stack.

Compliance

Standards we hold ourselves to

SOC 2 Type II

Saltbox is currently undergoing a SOC 2 Type II examination, evaluating our controls for security, availability, confidentiality, and processing integrity.

Contact security@saltboxmgmt.com for the latest status or to request our readiness documentation.

GDPR

Saltbox is fully compliant with the EU General Data Protection Regulation:

  • Data export — Administrators can export all organizational data on demand
  • Data deletion — Organization deletion permanently removes all associated data
  • Configurable retention — Set retention windows for chat history, audit logs, and meeting transcripts
  • Data minimization — We collect only the data necessary for platform functionality

Audit logging

Every security-relevant action is recorded in a comprehensive audit log scoped to your organization — authentication events, member changes, integration updates, data exports, Salesforce operations, and administrative actions.

Logs include contextual metadata and are retained for up to two years by default.

Security Testing

Tested on every change

Automated security test suites run on every code change before it reaches production. Our testing covers:

  • Data isolation — Verified separation between organizations and between projects within the same organization.
  • Identity security — Protection against email spoofing, session hijacking, and parameter tampering.
  • Authorization enforcement — Every protected endpoint is tested for unauthenticated access, role violations, and cross-tenant access attempts.
  • Input validation — Malformed inputs, injection attempts, and fabricated identifiers are handled safely without information leakage.
  • AI safety — Hallucination detection, tool authorization, and output validation.

Integration Security

Connecting your tools safely

Saltbox integrates with Salesforce, Jira, GitHub, Confluence, Google Drive, and more. Every integration follows the same principles.

Encrypted credentials

All integration secrets are encrypted at rest and decrypted only at the moment of use.

Least-privilege auth

We use OAuth 2.0 and app-level authentication wherever possible, avoiding personal access tokens.

Immediate revocation

Disconnecting an integration immediately removes stored credentials.

Audit trail

Every integration event (connect, disconnect, credential update) is recorded in your audit log.

Responsible AI

Built to assist, not to act alone

01

Human approval required

For any operation that modifies your Salesforce org, creates records, or takes external action.

02

Scoped access

The AI can only work within your authorized project and organization boundaries.

03

No autonomous outreach

The AI cannot send emails, post messages, or call external APIs without your explicit initiation.

04

Auditable

Every AI tool execution is logged with full context.

Incident Response

When something goes wrong

In the event of a security incident, we are committed to prompt investigation, timely customer notification (within 72 hours as required under GDPR), transparent communication about impact and remediation, and thorough root cause analysis.

Report a vulnerability

If you discover a security vulnerability, please report it to security@saltboxmgmt.com.

FAQ

Frequently asked questions

Where is my data stored?
Saltbox runs on cloud infrastructure hosted in the United States, including Vercel, Heroku, and Google Cloud. All data is encrypted in transit and at rest.
Does Saltbox train AI models on my data?
No. Your data is used only to deliver platform features. Our agreements with AI providers explicitly prohibit use of customer data for model training.
Can Saltbox employees access my data?
Administrative access is strictly controlled, logged, and auditable. Any support session involving customer data is recorded in the audit log.
What happens when I delete my organization?
All associated data is permanently and irreversibly deleted — conversations, projects, documents, integration credentials, and audit logs.
How long is my data retained?
Retention periods are configurable. Defaults are one year for chat history and transcripts, and two years for audit logs. Data beyond the retention window is automatically purged.
Can I export my data?
Yes. Administrators can request a full data export at any time.
Is Saltbox SOC 2 certified?
We are currently completing our SOC 2 Type II audit. Contact security@saltboxmgmt.com for current status and readiness documentation.

Have security questions?

Contact us for our compliance documentation, data processing agreement, or to schedule a security review.