/

CVE-2026-72801

Severity:

HIGH RISK

(7.5)

Credential Exposure (CWE-522) in SiYuan Publish Mode

CVE-2026-72801

Deferred

Vulnerability type

Category:

Insufficiently Protected Credentials

CWE:

CWE-522

Impact:

Offline Password Cracking and Notebook Confidentiality Loss

Severity:

HIGH RISK

(7.5)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

SiYuan publish mode may expose encrypted-notebook key-derivation material and wrapped data keys through endpoints that do not require authentication. An attacker who obtains these values could perform unlimited offline cracking against the notebook master password, especially when the password is weak or reused.

Affected context

The issue may occur in SiYuan deployments using publish mode with versions that expose encryption-related material through unauthenticated endpoints. The exposed information can include Argon2id salt and cost parameters, password verifiers, and wrapped notebook keys, allowing attackers to continue password-guessing attempts outside the application without online rate limits.

Exploit preconditions

  • Attack surface: Unauthenticated publish-mode endpoints
  • Required access: Network access to the exposed SiYuan service; no valid account may be required
  • Trigger point: Retrieval of encryption metadata and wrapped notebook key material
  • Why exploitation is possible: The disclosed values provide the inputs and verification context needed for unlimited offline master-password guessing

Why this happens

The vulnerability results from exposing credential-related cryptographic material through an unauthenticated interface. Although salts, cost parameters, verifiers, and wrapped keys may be intended to support encryption workflows, returning them without appropriate access controls weakens the protection boundary around the master password. This maps to CWE-522: Insufficiently Protected Credentials, because sensitive authentication or key-protection data is accessible to parties that should not receive it.

Potential impact

  • Attackers may conduct unlimited offline master-password cracking
  • Weak or reused passwords could be recovered
  • Encrypted notebook contents could be decrypted if the master password is obtained
  • Exposed key material could support targeted credential attacks
  • Confidentiality of published or locally stored notebook data could be reduced

General mitigation guidance

Mitigation generally focuses on preventing unauthenticated disclosure of encryption and credential-derived material.

Common defensive patterns include:

  • Restricting encryption metadata and wrapped keys to authenticated, authorized operations
  • Separating public publishing functionality from private key-management endpoints
  • Applying server-side authorization checks before returning sensitive cryptographic data
  • Reducing the amount of key-derivation and verification material exposed to clients
  • Using strong, unique master passwords that are not reused across services
  • Limiting publish-mode exposure through network controls, reverse-proxy access policies, and private deployment boundaries
  • Treating previously exposed notebook credentials or key material as potentially compromised and rotating them through an established recovery process

How teams detect issues like this

Organizations can assess this class of issue through:

  • Endpoint inventories identifying unauthenticated routes in publish mode
  • Automated tests that verify sensitive cryptographic metadata is inaccessible without authorization
  • Static and configuration analysis for routes returning salts, password verifiers, wrapped keys, or key-derivation parameters
  • Code review checks for missing authentication and authorization around encryption-related responses
  • Web-access log analysis for unusual requests to publish-mode key or notebook endpoints
  • Policy enforcement requiring explicit classification and access controls for credential-derived data

Sources

SourceLink
github.comgithub.com/siyuan-note/siyuan/security/advisories/GHSA-8x84-r2ff-h8pq
vulncheck.comvulncheck.com/advisories/siyuan-before-information-disclosure-via-encryption-key-material

Frequently asked

questions

It may be remotely exploitable when publish mode is reachable over a network and the affected endpoints disclose the material without authentication.

No account may be required for the initial retrieval. The critical prerequisite is network access to the exposed publish-mode endpoints.

Prevent unauthenticated access to encryption metadata, password verifiers, and wrapped keys, and ensure authorization is enforced before any such material is returned.

Classify cryptographic and credential-derived data as sensitive, test endpoint authorization in CI and security reviews, minimize client exposure, and use strong unique master passwords.

Find out what's already
exploitable in your codebase.