Integrating agentic risk signalling in trusted research environments: Automating VEX with Agent2Agent protocols and model context protocol (MCP) in SACRO and TREvolution pipelines

Download PDF
Open Webpage

Petar Radanliev, Kayvan Atefi, Omar Santos, Carsten Maple

Published: 01 Jan 2026, Last Modified: 12 Nov 2025Comput. Stand. Interfaces 2026EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Proposes agentic VEX generation using Agent2Agent and Model Context Protocol.•Embeds runtime traceability and exploitability logic in reproducible TRE pipelines.•Automates vulnerability filtering and digital signing in SACRO-compliant systems.•Supports federated reproducibility and schema validation for VEX artefacts.•Enhances AI supply chain assurance with context-aware risk signalling.