SIMULATION NOTICE: This document (audit, evaluation, response) is part of an AI-driven role-playing simulation conducted for project quality and governance testing. It does not constitute a formal legal or professional audit by any real-world entity.

1. Objective

The purpose of this report is to share the current security consideration status for Web/A L2 encryption and the Folio ecosystem with the Red Team, and to organize the discussion points for future detailed audits (v6 and beyond).

2. Current Status and Implementation Highlights

In the most recent phase, the following architectural advancements and policy decisions have been made:

2.1. Introduction of Static-Epoch Forward Secrecy (SEFS)

To achieve forward secrecy under the constraint of operating solely within a browser (fixed static files), the "Static-Epoch Forward Secrecy (SEFS)" model has been adopted.

  • Status: Prototype implemented.
  • Problem Solved: Successfully limited the risk of retroactive decryption of past communications to a specific window (e.g., within 24 hours) through periodic private key shredding on the server side.

2.2. Tiered Key Management Architecture (Defense in Depth)

Defined a hardware-backed key management model that does not rely on vulnerable storage areas like localStorage.

  • Tier 1 (Dev): File-based (current reference implementation).
  • Tier 2 (Standard): Utilization of physical Enclaves via WebAuthn / Passkeys.
  • Tier 3 (High): A model that links device keys with national IDs or other high-assurance credentials using Holder Binding technology.

2.3. Hybrid Considerations for "Folio Vault" and Cloud HSM

Due to browser storage constraints and the need for reliable backups, the following hybrid configuration is being considered:

  • Design Policy: Encrypt the entire Folio (AES-GCM) with a key derived from the end-user's Passkey and place it in the cloud as a "Vault."
  • Commercialization: For large-scale, high-assurance services, configurations where the Folio is hosted on Server-side HSMs / TEEs operated by the service provider are permitted.

3. Known Technical Challenges and Requests for Red Team Input

We request focused evaluation of the following points in future audits:

  1. Security of WebAuthn PRF Extension: The robustness of the method for deriving encryption keys using the PRF extension in browser environments where private keys for signing cannot be directly extracted from hardware.
  2. Trust Chain of Holder Binding: The degree of strength maintained by a Passkey once bound from a physical ID card if the device itself is compromised.
  3. Exploitation of Clock Skew in SEFS: Mitigation measures against replay attacks at Epoch boundaries or the risk of forced key selection via client-side clock tampering.

4. Future Plans

  • 2025 Q1: Release of the Folio CLI reference implementation integrating WebAuthn/PRF.
  • 2025 Q2: Definition of high-assurance profiles compliant with EUDIW ARF and PoC for server-side HSM integration.
  • 2025 Q3: Integration of the Pre-key Vending Machine to achieve "Tier 3 (True PFS)" for forward secrecy.

Based on these considerations, we look forward to continuous threat analysis and proposals for bypass methods from the Red Team.