Security at Quill

Data protection

Your data is yours. We never use customer data to train foundation models, and we never share it with third parties except the subprocessors required to deliver our service.

Encryption

All data is encrypted in transit (TLS 1.2+) and at rest (AES-256).

Credential isolation

Credentials for third-party integrations are never stored in Quill's database. A dedicated integration platform holds them under its own encryption and access controls.

Tenant isolation

Customer data is logically separated, with strict controls preventing cross-tenant access.

Data retention

You control what we retain and for how long. Deletion requests are honored within contractual timelines.

No foundation-model training

Your inputs, outputs, and agent activity are never used to train foundation models, by Quill or by any of our subprocessors. Quill uses only aggregated, de-identified usage data to improve its own product.

AI & agent governance

Quill's agents read data, trigger workflows, and act on behalf of your business, so governance is built into the runtime.

Isolated execution environments

Every customer automation runs in an isolated, short-lived serverless environment, fully separated from other customers and from Quill's control plane.

Scoped permissions

Every agent operates under least-privilege access, with explicit scopes for what it can read, write, and act on.

Human-in-the-loop controls

Sensitive actions can require human approval before they execute.

Auditable by design

Agent activity is captured end-to-end, giving admins visibility into every run.

Model provider controls

We work with model providers under no-training terms so your data is never used to train their models.

Infrastructure & access

Cloud infrastructure

Quill runs on enterprise cloud infrastructure with defense-in-depth controls across network, compute, and storage layers. Production and non-production environments are logically isolated in separate accounts.

Access controls

Internal access to production systems requires SSO and MFA, is scoped by role, and is logged and reviewed.

Monitoring

We continuously monitor infrastructure health, error rates, and latency. Critical alerts are routed directly to the on-call engineer.

Incident response

We maintain a documented Security Incident Response Plan with a dedicated on-call rotation. Internal and external parties can report security issues to security@tryquill.com.

Secure development

Code is reviewed before merge through protected branches, and only authorized, authenticated staff can trigger production deployments through our automated pipeline.

Vulnerability management

We continuously scan our infrastructure for known vulnerabilities. Findings are prioritized by severity and remediated through our standard change management process.

Automated backups

Databases are backed up automatically with point-in-time recovery. Backups are encrypted at rest.

Penetration testing

We conduct third-party penetration testing on a scheduled basis to independently validate our security controls.

Employee & endpoint security

Security training

All employees complete mandatory security awareness training and acknowledge Quill's security policies during onboarding.

Background checks

All employees undergo third-party background checks as part of the hiring process.

Device management

Employees use company-provided laptops only; no personal devices on production systems or sensitive data. All laptops have full-disk encryption, automatic screen lock, and active malware protection enforced.

Endpoint monitoring

Quill deploys a continuous endpoint compliance agent to all laptops, verifying encryption, password policy, screen lock, and OS patch level. Out-of-compliance devices trigger immediate alerts.

Access revocation

Access to all systems is revoked within 24 hours of an employee's last day, or sooner if warranted.