/

CVE-2026-85607

Severity:

HIGH RISK

(8.8)

Authorization Bypass (CWE-639) in Blinko Chat API

CVE-2026-85607

Deferred

Vulnerability type

Category:

Authorization Bypass / Insecure Direct Object Reference

CWE:

CWE-639

Impact:

Unauthorized data access and modification

Severity:

HIGH RISK

(8.8)

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

Blinko 1.8.7 contains an authorization bypass, commonly known as an insecure direct object reference (IDOR), in authenticated message and conversation procedures. An authenticated user who can supply another user's sequential message or conversation ID may read chat history, alter message content, or delete messages and conversations.

Affected context

The vulnerability occurs in Blinko's tRPC message and conversation procedures when they accept caller-supplied resource identifiers. Authentication is enforced, but authorization may not confirm that the referenced message or conversation belongs to the requesting account. Deployments are most exposed when identifiers are predictable and these procedures are reachable by ordinary authenticated users.

Exploit preconditions

  • Attack surface: Authenticated tRPC procedures for listing, updating, deleting, or clearing messages and conversations
  • Required access: A valid Blinko user account or session
  • Trigger point: A request containing a message or conversation identifier
  • Why exploitation is possible: Resource queries use caller-supplied identifiers without consistently enforcing ownership or tenant authorization

Why this happens

The underlying bug is an object-level authorization failure. The application authenticates the caller but treats possession of a valid resource identifier as sufficient authority to access or modify the corresponding record. This maps to CWE-639: Authorization Bypass Through User-Controlled Key, because user-controlled keys are used to select database objects without an ownership or access-scope check.

Potential impact

  • An authenticated user may read another user's AI chat history
  • Message content could be altered without the owner's authorization
  • Individual messages or complete conversations could be deleted
  • Confidentiality and integrity of user data may be compromised
  • Repeated abuse could cause broad account-level data loss

General mitigation guidance

Mitigation generally focuses on enforcing authorization at the data-access boundary rather than relying on authentication or predictable identifiers.

Common defensive patterns include:

  • Apply mandatory ownership, tenant, or policy predicates to every resource query and mutation
  • Derive the effective account or tenant from the authenticated session instead of trusting request fields
  • Use centralized authorization helpers or repository methods that require an access scope
  • Prefer opaque, non-sequential identifiers as defense in depth, without treating them as an authorization control
  • Return consistent not-found or access-denied behavior without disclosing whether another user's resource exists
  • Apply least privilege to database roles and service accounts
  • Add authorization tests for read, update, delete, and bulk-clear operations

How teams detect issues like this

Organizations can identify similar issues through:

  • Static analysis and code review searches for database lookups using request-supplied IDs without ownership predicates
  • Review of tRPC or API handlers that authenticate users but do not pass account or tenant context into repository queries
  • Tests that attempt cross-account access for every resource read and mutation
  • Runtime monitoring for sequential identifier enumeration, repeated authorization failures, and unusual cross-account access patterns
  • Policy enforcement requiring explicit authorization checks for object-level API operations
  • Automated security review tools that flag direct object references and missing access-control conditions

Sources

SourceLink
github.comgithub.com/blinkospace/blinko
github.comgithub.com/blinkospace/blinko/blob/1.8.8/server/routerTrpc/conversation.ts
github.comgithub.com/blinkospace/blinko/blob/1.8.8/server/routerTrpc/message.ts
github.comgithub.com/blinkospace/blinko/issues/1218
vulncheck.comvulncheck.com/advisories/blinko-1.8.7-cross-user-ai-conversation-read-and-write-via-message-trpc-router

Frequently asked

questions

Yes, within the application's authenticated API surface. Exploitation generally requires sending requests as a valid user and supplying identifiers for resources owned by another account.

No. The reported behavior can affect ordinary authenticated users when resource ownership is not checked by the relevant procedures.

Enforce ownership or tenant authorization in every resource query and mutation, using identity derived from the authenticated session rather than relying on a caller-supplied account or resource relationship.

Maintain centralized object-level authorization, use cross-account negative tests, review all identifier-based endpoints, and monitor for enumeration and anomalous access patterns.

Find out what's already
exploitable in your codebase.