Skip to main content
Review and revision metadata
Review Date: 2026-06-16
Reviewer: Director

previous version on gdrive

anDREa Security Manifesto

This Manifesto defines the core philosophical principles, design values, and structural paradigms that govern the myDRE platform ecosystem. It establishes the cultural and technical baseline for risk management under ISO 27001:2023 and NIS 2, clarifying how anDREa balances user enablement with institutional control.

This document undergoes mandatory review annually or immediately following significant technical, operational, or threat landscape shifts.


1. Core Security Value Statements

In engineering a trusted platform for organizations whose personnel must receive, process, analyze, and transfer highly sensitive data, anDREa adheres to a specific set of core values. We structure our priorities according to the following framework:


┌───────────────────────────────────────┐       ┌───────────────────────────────────────┐
│        WE STRONGLY VALUE:             │       │        OVER AND ABOVE:                │
├───────────────────────────────────────┤       ├───────────────────────────────────────┤
│ Operational Enablement                │ ───►  │ Total Risk Elimination                │
│ Autonomy and Local Responsibility     │ ───►  │ Rigid Bureaucratic Procedures         │
│ Foolproof, Frictionless Workflows     │ ───►  │ Total Feature Restrictions            │
│ Strict Containment and Isolation      │ ───►  │ Low-Cost Architectural Shortcuts      │
│ Incentivizing Good Behavior           │ ───►  │ Programmatic Barriers                 │
└───────────────────────────────────────┘       └───────────────────────────────────────┘

While we recognize the value in the items on the right, we prioritize the items on the left to achieve an optimal balance between security, compliance, and real-world research utility.

important

The Malicious Insider Boundary: While anDREa is highly diligent in blocking unauthorized processes and monitoring for anomalous behavior, the platform is not designed to restrict or block actions taken by explicitly authorized personnel acting with malicious intent. Managing insider threat vectors relies on organizational trust and external compliance governance.


2. Elaborated Security Principles

A. Operational Enablement vs. Absolute Risk Elimination

Completely locking down an environment to eradicate risk typically renders researchers unable to perform their work. When users face rigid programmatic barriers, they naturally migrate to unmanaged, highly insecure "shadow IT" alternatives that are invisible to the organization.

  • The Directive: "No" is not an acceptable operational answer for approved work packages. If an alternative, more secure path exists, the workspace maintains top-tier security. If a risk must be taken to complete the science, the least-vulnerable technical solution is implemented, using a strict Comply-or-Explain protocol to document the rationale.

B. Local Responsibility vs. Bureaucratic Procedures

Artificial scarcity or heavily delayed approvals for computing resources trigger artificial spikes in demand.

  • The Directive: By giving highly trained professionals on-demand, self-service access to resources paired with financial visibility, users inherently limit resource runtimes. Trusting researchers with immediate control over parameters (such as temporary inbound rules or de-allocation clocks) yields better security and lower operational costs than binding them to restrictive central processes.

C. Frictionless Workflows vs. Total Feature Restrictions

Target users for the myDRE environment possess elite intelligence and high technical dexterity. If a platform relies solely on restrictive rules, these users possess both the skills and the drive to uncover workarounds.

  • The Directive: anDREa focuses on eliminating the structural need for workarounds by delivering simple, foolproof, and smart workflows. Rather than trying to eliminate every conceivable edge-case vulnerability, the platform invests in easy to use functionality with logging as proof of compliance.

D. Strict Containment and Isolation vs. Low-Cost Shortcuts

From an individual researcher’s perspective, the statistical likelihood of an extreme security incident feels negligible. However, across a scaling enterprise, an unlikely event becomes an operational certainty over time, threatening millions of euros in recovery costs and permanent reputational damage.

  • The Directive: anDREa rejects low-cost infrastructure shortcuts. We assume that a security compromise will eventually occur. Therefore, our primary defense relies on total logical containment and cryptographic isolation between individual tenant subscriptions and workspaces, permanently neutralizing horizontal blast radiuses.

E. Incentivizing Good Behavior vs. Total Prevention

Enforcing rigid systems causes friction. The majority of platform users will operate safely if given the correct environment.

  • The Directive: We leverage a balanced approach to guide user behavior. We make secure functionalities easy and highly accessible (the carrot) while enforcing unalterable, comprehensive activity logging alongside a platform-wide ban mechanism for malicious actions (the stick). Knowing that every action produces an unalterable compliance trail deters bad behavior.

3. Threat Analysis & Mitigations Ledger

This matrix maps active threat vectors against the specific technical and organizational controls engineered into the myDRE platform architecture.

Threat Vector / Identified RiskImplemented Technical & Organizational Controls
Unauthorized Resource AccessAccess requires a valid, MFA-protected identity paired with an explicit workspace invitation from an Accountable or Privileged Member.
Orphaned Target AccountsAccountables retain self-service tools to immediately evict users from a Workspace. Stale identities are automatically isolated via system-enforced Access Review Policy.
Unauthorized Resource ModificationConfiguration privileges are restricted via RBAC to authorized Support Teams and pre-approved Workspace Owners (anDREas Role and Responsibilities Matrix).
Malicious Data IngressCode and files can only be uploaded by authorized Workspace members. Inputs are checked by native anti-malware tools.
Malware, Trojans, & RansomwareMitigated via active endpoint defense scanners, 30-day rolling 24-hour snapshot recovery arrays, and on-demand cold backups.
Phishing & Credential TheftUniversal enforcement of advanced Multi-Factor Authentication across all administrative and ingress boundaries.
Accidental Data ExfiltrationOutbound data transfers require an explicit, multi-stage approval workflow validated by a designated Privileged Member.
Unauthorized Data InterceptionMandatory cryptographic enforcement for all data streams in-transit (minimum TLS 1.2 / TLS 1.3 preferred).
Physical Infrastructure BreachFull cryptographic encryption applied to all data at-rest (AES-256) combined with Microsoft Data Center physical controls.
Stale VM Runaway CostsVMs defaulting to automated de-allocation profiles. Overriding auto-shutdown requires an active session state and explicit approval from a Privileged Member.
Unmonitored Internet IngressStrict outbound-only network configuration lists managed exclusively by Accountables and Privileged Members.
Remote Desktops (RDP) Brute-ForceVM access paths require secure portal authentication and Azure Bastion tunneling, eliminating public port exposure.
Administrative "Skeleton Key" ExploitsEnforced multi-factor authentication (MFA) and continuous administrative behavioral logging inside Azure Privileged Identity Management (PIM).
Active Compromise ResponseTenant CISOs can issue immediate user isolation requests. anDREa support executes a platform-wide account freeze while preserving forensic logs.
Infrastructure Denial of Service (DoS)Protected via continuous 24/7/365 telemetry monitoring, health heartbeats, and automated cloud edge throttling.
Insiders or Compromised AdminsInternal anDREa engineering access is locked behind strict least-privilege role bounds, mandatory MFA, and comprehensive audit trails.
Upstream Software VulnerabilitiesContinuous scanning via integrated CI/CD pipelines (SonarCloud/Dependabot) paired with automated operating system patching.

Security Architecture Reference Diagram

To see how these principles are structurally enforced at the infrastructure level, refer to the myDRE Highlevel Architecture.