AIIC AI Intelligence Centre

SOURCE-LINKED INTELLIGENCE

DeFiFlowBench: Benchmarking and Improving Safe Executability in Natural-Language DeFi Workflow Synthesis

arXiv · AI, language, vision and robotics · article · Sep 10, 2026 · UTC

A structurally valid DeFi workflow can still authorize a costly trade. We introduce DeFiFlowBench, a benchmark of 207 team-authored prompts for natural-language DeFi workflow synthesis. It measures graph coverage, configuration completeness, and declared safety predicates, then tests supported trade configurations on a local EVM. Direct, constrained, and few-shot prompting produce 14-19 unsafe held-out executions per configuration under a fixed 5% price-impact cap. A slippage bound derived from a quote does not prevent the price impact of the order itself. We propose Koan-Safe, which combines

Read original source ↗ Open in workspace

recordType
paper
region
Global

Evidence & attribution

First collected: 2026-09-20T19:02:05.452Z. This is not the publication date.