cs.CR · 2026-06-02 · No. 13

Cryptography and Security, 2026-06-02.

3 new papers in cs.CR. Titles, authors, abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →

01 — The papers

3 entries
  1. 01

    Ghost Tool Calls: Issue-Time Privacy for Speculative Agent Tools

    Bardia Mohammadi, Lars Klein, Akhil Arora, Laurent Bindschaedler

    cs.CR · cs.AI · cs.CL

    Tool-augmented language agents speculatively issue likely future tool calls to hide latency, but those calls leak inferred user intent to external services before the agent commits to the branch. Every external observer that received the call retains the disclosure after the agent abandons the branch. Timing is the issue, not authorization: no commit-time cleanup, read-only restriction, or access-control allow-list unsends what an observer...

    arxiv.org/abs/2606.02483 · PDF

  2. 02

    SeClaw: Spec-Driven Security Task Synthesis for Evaluating Autonomous Agents

    Hao Cheng, Changtao Miao, Tianle Song, Yin Wu, He Liu, Erjia Xiao, Junchi Chen, Xiaoyu Shi, Yichi Wang, Jing Yang,...

    cs.CR · cs.AI

    Autonomous LLM agents increasingly operate in stateful environments where they access tools, files, memory, and external services. While such capabilities enable complex real-world workflows, they also introduce security risks that are difficult to capture with existing evaluations. Current agent security benchmarks often rely on manually curated tasks, provide limited coverage of emerging threats, and focus primarily on final outcomes rather...

    arxiv.org/abs/2606.02302 · PDF

  3. 03

    AgentRedBench: Dynamic Redteaming and Integration-Aware Defense for LLM Agents over SaaS Integrations

    Hiskias Dingeto, Will Leeney

    cs.CR · cs.AI · cs.CL · cs.ET

    Indirect prompt injection in tool-use agents is a concrete production threat: LLM agents read from integrations (third-party services such as Gmail, Salesforce, or Jira accessed through tool calls) whose response content the user neither writes nor controls. Existing benchmarks under-measure the threat: most cover only a handful of integrations with the same attack payload replayed across runs, and open-source guards are trained on chat-style...

    arxiv.org/abs/2606.02240 · PDF

This edition is part of The Daily Abstract — cs.CR archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.

Colophon Set in Georgia, with system sans for interface chrome and a monospaced stack for code and paper identifiers. Sole accent: amber #D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.