LeanGuard: checking every tool call of an LLM agent against a policy written in Lean 4
The authors (Sankar, Ganguly, Singh and colleagues) propose LeanGuard, a runtime that checks every tool call an agent proposes against a policy written in Lean 4 before it reaches the environment; evaluation is fail-closed. On Tau-Bench with five Qwen models (0.8B–27B), unguarded agents executed at least 33 calls the policy forbids while LeanGuard executed none; on DTap, direct prompt-injection success fell from 57.9% to 5.9%. For calls denied because of argument values, proof-guided repair found an admissible edit for 88 of 252 denials.
Source: arXiv
Need an LLM chatbot or agent for your business?
Describe your need briefly; you will get clarifying questions and a scoped proposal before any work starts.
Related news
AI & LLM · 2026-11-04
Tencent releases Hy4: first vision and computer-use support, aimed at the top tier of open models
AI & LLM · 2026-11-03
Strategist: a meta-agent that evolves inference strategies as code, with 39% lower inference cost
AI & LLM · 2026-11-03