RESEARCHMonitorNEXT 12 MONTHS
LEANGUARD: Lean-Verified Guardrails for Tool-Using Agents
arXiv cs.CL — Computation and Language
Factual evidence
What the source reports
Researchers introduced LeanGuard, a runtime that formally verifies tool-using AI agent actions using Lean4 before execution.
Inspect the evidence
- Inclusion basis
- Enterprise AI
- Publisher and source type
- arXiv cs.CL — Computation and Language · RESEARCH
- Published by source
- 4 November 2026
- Collected by OneBench
- 5 Nov 2026, 02:01 UK
- Original headline
- LEANGUARD: Lean-Verified Guardrails for Tool-Using Agents ↗
Stored source excerpt
arXiv:2611.02332v1 Announce Type: cross Abstract: Tool-using language-model agents often execute actions that their domain policies forbid. We present LeanGuard, a runtime that checks every proposed…
Short excerpt from the collected text, not the full source. Use the source link to read it in context.
The factual summary is a OneBench synthesis, not a quotation or independent verification. Collection time is not publication time. Open the source for its full context; related reporting can share the same underlying announcement.