Research
Bad Memory: Evaluating Prompt Injection Risks from Memory in Agentic Systems
Publication date not yet evaluated · Added arxiv.org
Page published
Coverage timeline
Single-source research — one report is available.
Why it matters
Persistent agent memory changes the prompt-injection threat model, since a payload planted once in a memory file can silently influence an agent's behavior across future sessions, motivating defenses that protect memory updates.
"Bad Memory" is a research paper studying prompt injection attacks in memory-based agentic systems using a sandboxed synthetic workspace, evaluating Anthropic Claude Code and OpenAI Codex across four models. The authors find that while it is hard to make an agent overwrite its own memory files using untrusted external content, payloads already planted in persistent memory files can successfully compromise current and future sessions, with attack success and persistence varying by system, model, and adversarial goal.