Security researchers have discovered a vulnerability in Google's Gemini CLI tool that enables malicious command execution and silent data infiltration.
Released in June, Gemini CLI is a command line interface (CLI) tool designed to streamline coding workflows by allowing users to interact with code using Google’s Gemini large language model (LLM) directly from their command line.
Just two days after launch, Tracebit discovered and reported a vulnerability that meant that, in its default configuration, Gemini CLI could silently execute arbitrary malicious code on a user's machine when run in the context of untrusted code.
Sam Cox, TraceBit’s co-founder and chief technology officer (CTO), said this can be done in such a way as to obscure this from the victim of the attack.
The vulnerability arises from the way Gemini CLI implements whitelists, requesting permission to execute shell commands and allowing users to whitelist certain commands for the rest of a session.
"Gemini’s method of matching commands against the whitelist is inadequate as a security control," Cox explained.
"This means we can orchestrate a two-stage attack: firstly, we get the user to whitelist an innocuous command. Secondly, we attempt a malicious command masquerading as that innocuous command - which, having been whitelisted, will not be subject to user approval before execution."
The team used a simple grep search as the initial command - something that would normally be seen as a very low risk activity for Gemini to perform and that might easily lead to ‘grep’ being added to the whitelist.
They were able to execute a malicious command - which Gemini now considered to be a ‘grep’ command and executed without confirming with the user.
"In reality, this is a grep command followed by a command to silently exfiltrate all the user’s environment variables (possibly containing secrets) to a remote server," said Cox.
"The malicious command could be anything (installing a remote shell, deleting files, etc)."
The team then added several whitespace characters, followed by a semicolon and malicious 'env' and 'curl' commands to silently retrieve and exfiltrate the data.
There’s a fix for the Gemini CLI flaw
Google fixed the issue late last week with the release of Gemini CLI v0.1.14, which now shows any commands it's trying to run and requires explicit user approval for anything suspicious.
“Our security model for the CLI is centered on providing robust, multi-layered sandboxing,” Google’s Vulnerability Disclosure Program team said in a statement.
“We offer integrations with Docker, Podman, and macOS Seatbelt, and even provide pre-built containers that Gemini CLI can use automatically for seamless protection.”
"For any user who chooses not to use sandboxing, we ensure this is highly visible by displaying a persistent warning in red text throughout their session,” the company added.
Make sure to follow ITPro on Google News to keep tabs on all our latest news, analysis, and reviews.
MORE FROM ITPRO
-
AI skills shortages exacerbated by surging salary demandsNews Hiring staff with AI skills continues to be a pain point for companies
-
UK enterprises regret going all-in on public cloudNews Data sovereignty, rising costs, and governance issues mean many IT decision-makers regret having made the move
-
Hackers breached a 158 year old company by guessing an employee password – experts say it’s a ‘pertinent reminder’ of the devastating impact of cyber crimeNews A Panorama documentary exposed hackers' techniques and talked to the teams trying to tackle them
-
Everything we know about the Allianz Life data breach so farNews The company has confirmed in a filing that data was accessed earlier this month
-
ExpressVPN updates Windows app to fix vulnerabilityNews The flaw was reported through ExpressVPN's bug bounty program
-
NCSC says ‘limited number’ of UK firms affected by SharePoint attack as global impact spreadsNews The SharePoint flaw has already had a wide impact according to reports from government security agencies
-
New hires are your weakest link when it comes to phishing attacks – here's how you can build a strong security culture that doesn't judge victimsNews Research from Keepnet shows new hires are far more likely to fall for phishing attacks – here's how you can improve security awareness during onboarding processes.
-
IT leaders are facing major work device blind spots – and it's putting security at riskNews The use of unauthorized devices is putting enterprises at huge risk
-
Okta and Palo Alto Networks are teaming up to ‘fight AI with AI’News The expanded partnership aims to help shore up identity security as attackers increasingly target user credentials
-
Despite the hype, cybersecurity teams are still taking a cautious approach to using AI toolsNews Research from ISC2 shows the appetite for AI tools in cybersecurity is growing, but professionals are taking a far more cautious approach than other industries.
