Security & Trust
Enterprise trust, designed in.
How Digils protects your data, keeps you in control and limits what its AI can ever do.
Control plane
You control users, permissions, billing, agent policies and releases.
Execution plane
Agents run through queues with approvals, budgets, safe retries and recovery.
Data plane
Your records are isolated, audited, encrypted and backed up with tested restores.
Intelligence plane
AI runs through a controlled gateway with quality checks and governed learning.
Controls you keep
- Backend-enforced roles and attribute permissions
- Human approval before outreach and learning changes
- Spending limits per workspace and agent
- Emergency stop for all agent activity
- Full audit trail and instant access revocation
Digils will never
- Rewrite its own production code
- Grant itself new permissions
- Change your billing policies
- Access another customer's data
- Start messaging without authorization
Independent oversight
Digils Sentinel
A separate safety layer with policy checks, operational monitoring and emergency controls that act independently of the agents it watches.
Trust Center
Documents and disclosures.
Privacy Policy
How we collect, use and protect personal data.
ReadTerms of Service
The agreement for using Digils.
ReadCookie Policy
Why this website uses no tracking cookies.
ReadSubprocessors
Third parties that process data on our behalf.
ReadData residency
Where customer data is hosted and backed up.
ReadAI provider disclosure
Which models we use and how your data is handled.
ReadCertifications such as SOC 2 or ISO 27001, uptime commitments and regulatory attestations will appear only once independently established.
Security policy
How we keep Digils secure.
Website security
- HTTPS everywhere, with HTTP Strict Transport Security.
- A strict Content Security Policy: no inline code and no third-party scripts.
- Protection against clickjacking, content sniffing and cross-site attacks through security headers.
- Form submissions are rate-limited, validated on the server and delivered directly to our team by email; this website keeps no database of visitor data.
Platform security by design
- Tenant isolation, so one customer can never see another customer’s data.
- Backend-enforced role- and attribute-based access control.
- A complete audit trail of agent and staff actions.
- Encryption in transit, encrypted backups and tested restores.
- Human approval before outreach, spending limits and an emergency stop for all agent activity.
Independent certifications such as SOC 2 or ISO 27001 will be listed here only once they are achieved.
People and access
- Least-privilege access to production systems, reviewed regularly.
- Multi-factor authentication for administrative access.
- Confidentiality obligations for everyone with access to customer data.
Incident response
If we detect a security incident, we investigate, contain it and restore normal operation. Where personal data is affected, we notify affected customers and the relevant authorities within the time limits the law sets.
Responsible disclosure
If you believe you have found a vulnerability, email contact@digils.ai with the subject "Security", including steps to reproduce it. Please do not access other people’s data, disrupt our services or use social engineering. We will acknowledge your report within three business days and keep you informed. We will not take legal action against good-faith research that follows this policy. Our contact details are also published at /.well-known/security.txt.