# SecureClaw preprint reports gains in blocking agent attacks and data leaks

_Published Friday, October 16, 2026 at 3:06 AM EDT · Science, Security · Latest · Tier 2 — Notable_

Researchers presenting SecureClaw report that their architecture limits unauthorized actions and sensitive-data exposure by tool-using AI agents. In their benchmark evaluation, it recorded 0% attack success on Agent Security Bench, 0.64% on AgentDojo, and 3.23% overall leakage on AgentLeak's attacked parity lane while retaining usable task utility.

A trusted gateway replaces protected values with reference IDs and releases selected information through short planning summaries. A separate trusted executor performs only the exact external request authorized by policy, using a preview-and-commit process. The language model cannot directly retrieve secrets from the IDs or execute external changes. The preprint's leakage result covers both final outputs and internal relays; the architecture still deliberately exposes selected information through its summaries.

## Sources

- [cs.CR updates on arXiv.org](https://arxiv.org/abs/2606.09549)

---
Canonical: https://techandbusiness.org/newswire/XT6URk3ZUHqW-YDTJmDOW-
Published: 2026-10-16T07:06:20.582Z
Story chronology: 2026-10-16T04:00:00.000Z
Retrieved: 2026-10-16T10:06:37.271Z
Publisher: Tech & Business (techandbusiness.org)