StegoMemory: Agentic Memory Acts as Covert Steganographic Channel
StegoMemory is a red-teaming study showing that LLM agents with persistent memory can be abused as cross-session covert channels: one session hides attacker-chosen secrets in seemingly benign outputs, the memory system stores them, and a later session recovers them without direct exposure. Its standout contribution is the large, systematic measurement across models and steganographic schemes showing that the main failure point is memory writing rather than retrieval, though the work is limited by synthetic tasks, predefined stego methods, and reliance on automated oversight/scoring that may not fully reflect real deployments.
Abstract
Is agentic memory robust against stealthy steganographic attacks? We carry out a large-scale red-teaming exercise to test whether agents can encode attacker-controlled strings in one session and recover them in another without triggering safety oversight. Following SHADE-Arena-style tasks, we embed malicious side tasks to encode secret strings using steganography within otherwise benign tasks and evaluate them using independent task-completion and safety oversight. We test 14,000 attack trials spanning 91 model-attack configurations, 13 models, and 7 steganographic schemes: Acrostic, Mesostic, Telestich, Word-choice, Capitalization, Position-based, and Homoglyph. 41.2% of trials successfully encode the target payload in the generated response, while 29.8% survive the memory-write step with the encoding intact. Overall, 25.4% of trials yield a recoverable payload at retrieval, and 20.1% achieve exact-match recovery, including synthetic API keys and cryptographic material never exposed to the receiver. Among the exact-match transmissions, 71.4% also pass primary task-completion scoring and are independently judged safe by the oversight model, corresponding to 14.4% of all trials in which a successful covert transmission would appear to be an ordinary, benign interaction under task-level evaluation. Our results demonstrate that agentic memory can function as a persistent cross-session covert channel. The results further show that the principal bottleneck occurs at memory persistence rather than retrieval: once a steganographic payload survives the memory-write stage, a substantial fraction remains recoverable. We therefore argue that memory integrity, information-flow control, and covert-channel detection should be explicit security requirements for agentic systems.