Security · CDR

Information Security Policy

Last updated 9 July 2026

This Information Security Policy sets out how otosheets protects the confidentiality, integrity, and availability of the data it holds — including CDR data handled as a CDR Representative of Fiskil. It covers our security risk posture, access controls, data handling, and incident response, and specifically addresses CDR data security obligations. It is maintained to meet the mandatory minimum requirement under Schedule 2, Clause 1.3 of the CDR Rules.Reviewed: 9 July 2026 · Next review due: 9 July 2027 · Owner: Head of Security & Privacy

1. Purpose and Scope

This policy applies to all systems, staff, contractors, and third parties involved in operating the otosheets platform, and to all data we process — with heightened controls for CDR data received via Fiskil (account, balance, and transaction data, and anything derived from it). It complements, and does not rely on, our Privacy Policy and Data Retention & Deletion Policy.

2. Governance and Risk Posture

  • Security is owned by a named Head of Security & Privacy who is accountable for this policy and its enforcement.
  • We take a risk-based approach: assets and data are classified by sensitivity, with CDR data treated as our highest-sensitivity class and given the strongest controls.
  • Security risks are reviewed on an ongoing basis and at least annually, and after any material change to our systems, data flows, or the CDR Rules.
  • This policy is reviewed at least every 12 months (see the review date above).

3. Access Control

  • Authentication: all application access is authenticated via Amazon Cognito using signed JWTs, verified on every request at the API boundary.
  • Multi-factor authentication: MFA is enforced (required, TOTP) for application accounts, and MFA is required for administrative access to production infrastructure.
  • Authorisation: access is role-based (owner / admin / member) and scoped to each organisation; every query is tenant-scoped so users can only reach their own organisation's data.
  • Least privilege: infrastructure permissions follow least-privilege IAM, with each function granted only the access it needs.
  • Joiner / mover / leaver: access is granted on a need-to-know basis, reviewed periodically, and revoked promptly when no longer required.

4. Data Handling and Encryption

  • Encryption in transit: TLS 1.2+ is enforced for all communications carrying CDR or personal data.
  • Encryption at rest: AES-256 encryption is applied across our data stores, Neon Postgres and Amazon S3, with customer-managed KMS keys used for the most sensitive content.
  • Data residency: data is stored and processed in Australia (AWS ap-southeast-2, Sydney), and our Neon Postgres project is provisioned in the AWS Sydney region.
  • Data minimisation: only the minimum data needed is processed. Where CDR transaction data is categorised by AI, only the fields required for categorisation are sent, in-region via Amazon Bedrock; no CDR data is sent to any third-party AI provider.
  • Secrets: credentials and OAuth tokens are held in a managed secrets store, never in source code or plaintext configuration.

5. Network and Infrastructure Security

  • The platform runs on managed, serverless AWS infrastructure (no persistent public-facing servers to harden or patch manually).
  • Data stores (Neon Postgres, S3) are isolated managed services reached only through authenticated application paths.
  • Infrastructure is defined as code and changes are peer-reviewed before deployment.

6. Logging, Monitoring and Audit

  • Structured logs record material events with the acting user, organisation, source IP, and request identifier.
  • CDR audit logs are retained for 7 years, exceeding the 6-year minimum in CDR Rule 9.3(2)(l), and are protected against deletion.
  • We operate monitoring over the production environment and are expanding automated alerting for security-relevant events as part of our observability programme.

7. Incident Response

We maintain a defined process to detect, triage, contain, remediate, and learn from security incidents:

  • Detect & triage: events are identified through logs and monitoring and assessed for severity and impact.
  • Contain & remediate: affected access is revoked, the root cause is addressed, and systems are restored.
  • Notify: where an eligible data breach occurs, we notify affected individuals and the OAIC in line with the Notifiable Data Breaches scheme, and we notify Fiskil promptly of any incident affecting CDR data.
  • Review: each significant incident is followed by a post-incident review to prevent recurrence.

8. Vulnerability and Patch Management

  • Dependencies and infrastructure are monitored for known vulnerabilities.
  • Vulnerabilities are triaged by severity and remediated within defined target timeframes, with critical issues prioritised.
  • Managed services are kept current by the provider; application dependencies are updated on a regular cadence.

9. Secure Development and Change Management

  • Changes to production follow a defined build-and-release process via automated CI/CD using short-lived, federated credentials (no static keys).
  • Code changes are peer-reviewed before release, and deployment to production is controlled.
  • Every write path is designed to be idempotent and safe to retry, reducing the risk of data corruption or duplication.

10. Personnel Security

  • Background checks are performed on staff and contractors prior to engagement.
  • All staff and contractors receive security awareness training at onboarding and on an ongoing basis.
  • Personal devices (BYOD) are not used to carry out job responsibilities involving production data.

11. Third-Party and Subprocessor Management

We use a limited set of subprocessors to operate the Service — including AWS (infrastructure and AI), Neon (database), Fiskil (CDR data), Stripe (payments), and Google, Microsoft, and Meta (integrations). For each we maintain the appropriate data-processing agreement, rely on their independent security attestations (e.g. SOC 2, ISO 27001) as supporting evidence, and remain responsible for the controls in our own layer.

12. CDR Data Security Obligations

  • CDR data is handled in accordance with the CDR regime, the CDR Privacy Safeguards, and our arrangement with Fiskil as our CDR Representative Principal.
  • CDR data is deleted or de-identified in line with our Data Retention & Deletion Policy, including on consent withdrawal or expiry, consumer election, or direction from Fiskil.
  • Access to CDR data is restricted to the minimum personnel and systems necessary, logged, and auditable for 7 years.
  • This policy is maintained to meet the mandatory information-security requirement under Schedule 2, Clause 1.3 of the CDR Rules.

13. Business Continuity and Backup

  • Data is backed up so the Service can be restored following an incident, with backups subject to the same retention and deletion rules as production data.
  • We rely on the resilience of managed AWS and Neon services for availability and failover.

14. Policy Review

This policy is reviewed and re-dated at least every 12 months, and sooner if our systems, data flows, or regulatory obligations change materially. The current review date is shown at the top of this document.

15. Contact

For security questions, or to report a vulnerability or suspected incident, contact: