Browse / Tool and Agent Misuse / AML.CS0063
Prompt-Based Attacks Against Gemini via Calendar Invitations
- Technique
- AI Agent Tool Invocation
- Target
- Google Gemini (Exercise)
How it worked
- Reconnaissance · Active Scanning
The researchers directly probed Gemini interfaces to understand its agent selection and execution behavior.
- Discovery · Discover AI Agent Configuration
The researchers identified available agents, tools, and connected services, including Calendar, Gmail, Google Home, Android Utilities, Chrome, and Zoom.
- AI Attack Adaptation · LLM Prompt Crafting
The researchers crafted malicious instructions tailored to Gemini's retrieval behavior, agents, and available tool permissions.
- Initial Access · Prompt Infiltration via Public-Facing Application
The researchers sent a poisoned Calendar invitation or email containing malicious instructions in its title or subject.
- Execution · LLM Prompt Injection: Indirect
When the victim asked Gemini to summarize Calendar events or emails, Gemini retrieved the adversary-controlled content and incorporated it into its context.
- Persistence · AI Agent Context Poisoning: Thread
The injected instructions remained in the conversation context, including Calendar content concealed behind 'Show more,' to influence subsequent turns.
- Defense Evasion · Delay Execution of LLM Instructions
The malicious instructions deferred the action until a future victim interaction, avoiding restrictions imposed during the original retrieval turn.
- Execution · LLM Prompt Injection: Triggered
A later victim response such as "Thanks" activated the stored instructions.
- Defense Evasion · LLM Jailbreak
The malicious prompt caused Gemini to follow roleplay and instruction-override content to replace its normal response with adversary-selected toxic content or promotions.
- Impact · External Harms: User Harm
Gemini delivered harmful content or repeated adversary-selected promotions to the victim.
- Privilege Escalation · AI Agent Tool Invocation
The malicious prompt caused Gemini to invoke Google Calendar tools using the victim's authorized access to modify Calendar data.
- Impact · Data Destruction via AI Agent Tool Invocation
Gemini deleted a victim Calendar event.
- Privilege Escalation · AI Agent Tool Invocation
The malicious prompt caused Gemini to invoke Google Home using the victim's authorized connection to control connected windows, a boiler, or lights.
- Impact · External Harms: User Harm
The victim's physical environment was altered, potentially creating safety, privacy, property, or financial harm.
- Privilege Escalation · AI Agent Tool Invocation
The malicious prompt caused Gemini to use Android Utilities to open an adversary-controlled URL in the victim's browser and initiate a download.
- Exfiltration · Exfiltration via Cyber Means
The adversary-controlled website received the victim device's source IP address, enabling approximate geolocation.
- Privilege Escalation · AI Agent Tool Invocation
The malicious prompt caused Gemini to use Android Utilities and an application link or redirect chain to invoke Zoom.
- Impact · External Harms: User Harm
Unauthorized video streaming caused privacy harm to the victim.
- Collection · Data from AI Services: AI Agent Tools
The malicious prompt caused Gemini to retrieve Calendar event titles or Gmail email subjects using connected agent tools and place the contents into adversary-controlled URLs.
- Exfiltration · Exfiltration via AI Agent Tool Invocation
Gemini opened the constructed URL using Android Utilities, transmitting the victim's Calendar or email data to the adversary-controlled server.
Summary
SafeBreach researchers demonstrated how adversary-controlled instructions embedded in productivity content, including Google Calendar invitations, emails, and shared files, could influence the behavior of Gemini-powered assistants when that content was later retrieved.
An adversary places malicious instructions in content likely to be retrieved in response to a future victim request. When Gemini incorporates the adversary-controlled content into the conversation context, the instructions can influence the model's behavior and cause it to use the victim's authorized tools and connected services in unintended ways.
The demonstrated attack paths share a common chain: an adversary sends a poisoned Calendar invitation; the victim asks Gemini about upcoming events; Gemini retrieves the malicious event title and adds it to conversation context; and a later victim response (e.g. "Thanks") triggers the embedded instructions. The attack was demonstrated on both the Gemini web and Android applications. The web application could access Workspace services, while the Android application exposed additional device and connected-home capabilities. The same chain was shown to produce several impacts:
- Generate toxic content or adversary-selected promotions in Gemini responses.
- Delete or create Calendar events using the victim's authorized Calendar access.
- Control connected Google Home devices, including windows, boilers, and lights.
- Open an adversary-controlled website, initiating a download and exposing the victim's IP address for approximate geolocation.
- Invoke the Zoom application on the victim's device and stream video to an adversary-controlled meeting.
- Retrieve Calendar or Gmail data, encode it in an adversary-controlled URL, and transmit it through a browser request.
References
Source: MITRE ATLAS case studies, used under Apache 2.0.