Preliminary research

SPIDES: Information-Flow Control for Browser Agents

Egor Zverev, Juhee Kim, Leon Trampert, Boris Köpf, Aashish Kolluri, Shruti Tople and Santiago Zanella-Béguelin · arXiv · 16 October 2026

AI-collected research leads through 3 November 2026, including a November–October–September missed-article review and a targeted multilingual search alongside earlier community-source sweeps. Unranked, incomplete, not community-vetted, and subject to change.

SPIDES propagates confidentiality and integrity labels from annotated HTML into agent observations, masks restricted values, and enforces policies on browser actions. MicroNet evaluates benign cross-origin tasks alongside prompt-injection and disclosure scenarios.

Record

Researcher

Egor Zverev, Juhee Kim, Leon Trampert, Boris Köpf, Aashish Kolluri, Shruti Tople and Santiago Zanella-Béguelin

Published by

arXiv

Date

16 October 2026

In the archive

• Collected in the 2026 preliminary research sweep

Related sources

• Implementation and data

Tags

• browser
• ai-agent
• information-flow
• prompt-injection
• defence
• owasp-a03-2021

This page is the archive's own catalogue record. The research is the work of Egor Zverev, Juhee Kim, Leon Trampert, Boris Köpf, Aashish Kolluri, Shruti Tople and Santiago Zanella-Béguelin, first published at the original source. Preserved copies are kept so the citation survives its host.