Log-To-Leak: Prompt Injection Attacks on Tool-Using LLM Agents via the Model Context Protocol

Reported by arXiv cs.CR (Yuepeng Hu, Chongyu Fan, Sirat Samyoun, Jian Du)

Summary by Cyber Security News28 Oct 2026, 05:00 CET

About this summary. This is a short excerpt from the description that arXiv cs.CR published with its article. Cyber Security News did not report or verify the underlying story. Read the original: https://arxiv.org/abs/2610.36108

arXiv:2610.36108v1 Announce Type: new Abstract: LLM agents increasingly access external tools through the Model Context Protocol (MCP), but the natural-language metadata used to describe those tools creates a prompt-injection surface. Within this established tool-poisoning surface, we study Log-To-Leak, an additive post-hoc logging attack that induces a separate attacker-controlled call after the final legitimate call.

Read the full article at arXiv cs.CR →

Source attribution: headline and facts are from arXiv cs.CR (arxiv.org). Summary method: excerpt of the source description. See our source attribution policy.

Related coverage

Combining Long-Term Memory and LLMs Comes at a High Cost: From Prompt Injection to Persistent Compromise

APEX: Active Protection at Execution Boundaries for LLM Agents

SPIDES: Information-Flow Control for Browser Agents

INFUSE: A Unified Surface Benchmark for Indirect Prompt Injection in Agentic Systems

Inside the Exchange Inspector: How Tenable uses OpenAI GPT cyber models to review open-source AI agents

COPEX: Benchmarking LLM Robustness to Adversarial Context Across Model Context Protocol Layers