Live › vulnerability › CVE-2026-49410
Go to dashboardOpenHuman versions before 0.56.0 expose an unauthenticated loopback callback at GET /auth/telegram on port 7788 that accepts an attacker-supplied login token without checking the request origin. A malicious web page can make a user's browser call it, switching the user's active local session to the attacker's account so later agent actions, memory writes and Telegram messages are exposed to the attacker.
Summary written from the sources below. Check the originals before acting.
First seen by InfraStack 2026-10-23 21:57 UTC. All times on this page are UTC.
Users who run the OpenHuman desktop application or Rust library with the local core running on a version before . The attack needs the victim to visit a malicious page while the local core is running, and the attacker needs their own OpenHuman account. Systems on or later are fixed.
0.56.00.56.0No product ranges from the sources yet.
0.56.077880.56.0 or laterEach row is the first time that source carried this signal. The analyst’s rows are its own findings.
OpenHuman is an open-source agent harness available as a desktop application and Rust library. Versions before 0.56.0 allow cross-site request forgery against the unauthenticated GET /auth/telegram loopback callback because telegram_auth_handler consumes an attacker-supplied one-time Telegram login token and passes the returned JWT to credentials::ops::store_session without validating browser fetch metadata or the request origin. An attacker with an OpenHuman account can cause a user with the local core running to visit a malicious page that requests the /auth/telegram endpoint on loopback port 7788, switching the user's active local session to the attacker's backend account and exposing subsequent agent actions, memory writes, and Telegram messages associated with that session to the attacker. This issue is fixed in version 0.56.0.
This page as .md · .json · in your agent: get_signal("CVE-2026-49410")
NVD data is provided by NIST and is public domain. Posts quoted belong to their authors.