Tool-using LLM agents interact with the world through actions that persist state in artifacts (e. g.
Summary generated by The Flow from the publisher's feed. The full article lives at Hugging Face Trending Papers.
arXiv:2606. 09084v1 Announce Type: cross Abstract: Tool-using LLM agents interact with the world through actions that persist state in artifacts (e.
arXiv:2510. 01359v2 Announce Type: replace-cross Abstract: Code-capable large language model (LLM) agents are embedded in software engineering workflows where they can read, write, and execute code, raising "jailbreak" stakes beyond text-only settings.
arXiv:2607. 00481v1 Announce Type: cross Abstract: Jailbreak attacks remain a critical threat to the safe deployment of large language models (LLMs).
Tool-using agents consume external data from sources with different levels of trust, yet tool responses rarely identify who produced each component or what it should convey. We show that this gap enables state-corruption attacks, in which attacker-controlled content makes environmental claims beyond the informational authority of its response component and corrupts the agent's perceived environment, making the resulting action appear justified to existing guardrails.
arXiv:2608. 12789v1 Announce Type: cross Abstract: Tool-using agents consume external data from sources with different levels of trust, yet tool responses rarely identify who produced each component or what it should convey.
arXiv:2606. 18356v1 Announce Type: cross Abstract: Tool-using language-model agents introduce security failures that go beyond unsafe text: they can disclose protected objects, write persistent memory, send messages, modify databases, or trigger harmful code and tool effects.