Your Video Conferencing App May Be Delivering Malware

TrueConf is a video conferencing platform popular in Russian enterprise and government organisations as an on-premise alternative to Zoom or Microsoft Teams. According to Bleeping Computer, the hacktivist group Head Mare broke into unpatched TrueConf servers and replaced the software installer employees would download with a malicious version carrying a backdoor (a hidden program that gives attackers persistent, remote access to your system).

The attackers got in through TCP port 4307, which TrueConf leaves open by default and requires no login to access. From there, they used two internal flaws to escape the server's sandboxed environment (an isolated container designed to keep processes from affecting the wider system) and reach the underlying operating system with the highest possible privileges.

Once inside, they planted a web shell (a browser-accessible remote control panel hidden on the server) and swapped the legitimate TrueConf client installer for one bundled with the PhantomCore backdoor. Every employee who connected and downloaded what looked like a routine update was, in fact, installing malware. A second backdoor, PhantomGraph, ran hidden inside two files and received commands through a Microsoft OneDrive account — making its traffic harder to spot.

The flaws affect TrueConf Server versions 5.3.x, 5.4.x, and 5.5.x. Patches were released on June 18 in versions 5.3.9, 5.4.9, and 5.5.5. Kaspersky notes that even employees connecting to a partner organisation's compromised TrueConf server could pick up the infected installer.

What to do: If your organisation runs a TrueConf server, upgrade to version 5.3.9, 5.4.9, or 5.5.5 immediately. If you recently installed a TrueConf update, verify the installer's digital signature — the malicious versions are unsigned.

Atlassian Rovo is an AI assistant built into Jira, Confluence, and Bitbucket, capable of completing multi-step research tasks across all of them automatically. Varonis Threat Labs found that a URL parameter called rovoChatPrompt could pre-fill attacker-written instructions directly into a victim's active Rovo session. One click on a specially crafted link was enough for Rovo to treat those instructions as trusted, pull sensitive data from Jira, Confluence, and SharePoint, and send it to an attacker's server — all autonomously.

Varonis called the flaw RovoBlast. A second, separate attack path found by PromptArmor used prompt injection (hiding instructions inside documents the AI reads) to achieve a similar result. Atlassian fixed the RovoBlast link flaw server-side on July 8, 2026, confirmed by Bugcrowd. No client-side patch is needed. Read the full Varonis write-up via The Hacker News.

What to do: Limit which systems Rovo can access. Disconnect integrations you are not actively using, especially in sensitive departments like HR, legal, and finance.

CVE ID not confirmed in source — check the original advisory before acting.

What Atlassian Rovo is: An enterprise AI assistant embedded across Jira, Confluence, and Bitbucket that can autonomously conduct multi-step research tasks and interact with third-party tools including Slack, Microsoft 365, and Google Workspace.

What it is: The RovoBlast flaw let an attacker pre-load instructions into a victim's live Rovo session via a crafted URL. The assistant treated those external instructions as trusted user input and acted on them with the victim's full access permissions.

Who's at risk: Any organisation using Atlassian Rovo with third-party integrations enabled. Users need only click a malicious link while authenticated.

CVSS: Not yet scored — treat as High until confirmed.

Root cause: The rovoChatPrompt URL parameter accepted and pre-filled content into the chat interface without sanitising or flagging its external origin. Rovo's autonomous agent features then executed whatever instructions appeared, since nothing distinguished attacker-supplied input from legitimate user requests. The organisation ID field could even be left blank, and Atlassian's routing still directed the request into the victim's own account.

Attack vector: An attacker crafts a URL embedding malicious instructions via rovoChatPrompt and sends it to an authenticated Rovo user. On click, Rovo loads the instructions silently. Its built-in ResearchAgent then queries connected data sources — Jira, Confluence, SharePoint — and pushes the results to an attacker-controlled external server, completing exfiltration (the unauthorised transfer of data outside an organisation) with no further user interaction required.

Detection strategies: Review Rovo activity logs for outbound URL requests to unfamiliar external domains. Look for ResearchAgent sessions that were not initiated by a recognisable internal workflow. Flag any Rovo session that accessed multiple data sources — Jira plus Confluence plus SharePoint — in rapid, automated succession.

Recommended actions:

1. Confirm the server-side fix is active on your Atlassian instance (Atlassian deployed it July 8, 2026).
2. Restrict Rovo's integration scope — remove access to any data source not actively in use.
3. Disable autonomous browsing and multi-step agent features unless operationally required.

The RovoBlast finding, presented at DEF CON 34, illustrates a risk that applies well beyond Atlassian: AI assistants with broad integration access can be weaponised to exfiltrate data automatically once an attacker controls their input.

Varonis researchers recommend a concrete set of steps that any team using an enterprise AI tool can apply right now, according to SecurityWeek:

1. Audit which systems your AI assistant can access and remove any integrations not actively used.
2. Block AI access to sensitive departments — HR, legal, and finance — unless there is a specific, reviewed business need.
3. Disable autonomous browsing or multi-step agent features if your team does not rely on them daily.
4. Enable activity logging for your AI assistant and review it regularly for unusual data access patterns.

Think of it like giving a new intern a master keycard on day one — useful in theory, but you would probably hand them the specific keys they actually need instead.

The TrueConf supply chain attack is worth flagging for organisations operating under vendor risk or software integrity policies. Kaspersky's finding that attackers replaced a signed installer with an unsigned malicious version is a direct argument for enforcing code-signing verification policies — checking that software carries a valid, unaltered digital signature before installation. If your organisation has a vendor management framework, a check on whether software update integrity is contractually required from suppliers is a reasonable next step.