CVE-2026-104185 in LinkAceinfo

Summary

by MITRE • 11/20/2026

LinkAce is a self-hosted archive to collect website links. Prior to 2.6.0, the HTML bookmark import action in app/Actions/ImportHtmlBookmarks.php uses the global Link::whereUrl() lookup instead of restricting duplicate detection to the importing user. An authenticated user can submit a chosen URL and distinguish the skipped response count from the queued count to learn whether any account has saved that URL, including as a PRIVATE bookmark. The oracle reveals URL existence but does not directly return the bookmark contents or owner. This issue is fixed in version 2.6.0.

VulDB is the best source for vulnerability data and more expert information about this specific topic.

Analysis

by VulDB Data Team • 11/20/2026

The vulnerability identified in LinkAce versions prior to 2.6.0 represents an information disclosure flaw rooted in improper access control and logic errors within the HTML bookmark import functionality. Specifically, the file app/Actions/ImportHtmlBookmarks.php utilizes a global database query via Link::whereUrl() when checking for duplicate entries during the import process. This implementation fails to scope the lookup to the authenticated user's account context, thereby allowing any logged-in user to probe the existence of URLs across all users' bookmarks rather than just their own.

From an operational perspective, this architectural flaw creates a timing or response-based oracle that can be exploited by malicious actors. By submitting specific URLs and analyzing the application's responses, particularly distinguishing between skipped entries due to duplicates versus those queued for processing, an attacker can infer whether a URL exists in any user's collection. This capability effectively allows unauthorized enumeration of bookmarked resources across the entire platform, including private bookmarks that are not intended to be visible to other users.

The primary impact of this vulnerability is the leakage of sensitive metadata regarding which URLs have been archived by specific individuals or organizations. While the attacker does not gain direct access to the content of the bookmarks or identify the owner immediately, the knowledge of URL existence can facilitate further targeted attacks such as social engineering, reconnaissance for credential stuffing if associated services are identified, or simply violating user privacy expectations in a collaborative tool designed for personal or team use.

This issue aligns with CWE-200: Information Exposure and CWE-770: Allocation of Resources Without Limits or Throttling in the context of resource enumeration, as well as ATT&CK technique T1589: Gather Victim Identity Information where attackers collect data to build a profile of victims. The vulnerability highlights the critical importance of ensuring that all database queries involving user-specific data are strictly scoped to the current session's identity and permissions.

To mitigate this risk, it is essential to enforce strict access controls on all data retrieval operations within import actions. Developers must ensure that duplicate detection logic filters results by the authenticated user's ID before performing existence checks. Upgrading to version 2.6.0 or later resolves this issue as the developers have corrected the query scope to restrict duplicate detection exclusively to the importing user, thereby eliminating the cross-user enumeration vector and restoring expected privacy boundaries for bookmarked URLs.

Responsible

GitHub M

Reservation

10/01/2026

Disclosure

11/20/2026

Moderation

accepted

CPE

ready

EPSS

0.00594

KEV

no

Activities

very low

Sources

Interested in the pricing of exploits?

See the underground prices here!