mlsanityData to 7 Oct 2026

Symbolic Temporal Supervision of LLM Agents Using Contracts

arXiv 2609.18128 ↗First posted 16 Sep 2026cs.AI cs.LOcited by 0

On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks

Abstract (arXiv)

Large language model (LLM) agents augmented by tools can automate complex, multi-step tasks, such as web navigation, code generation, and workflow orchestration, by acting on external systems through tool calls. However, hallucinations, distributional instability, and adversarial manipulations in LLMs, and the irreversible consequences of certain tool calls can lead to harmful outcomes. Existing safeguards either grade recorded trajectories post hoc with stochastic LLM judges or block unsafe actions one call at a time, and no single deterministic artifact supports both roles. We present ContrAgent, a contract-based framework for symbolic temporal supervision of LLM agents. ContrAgent captures an agent's behavior as a sequence of tool calls and formalizes it as a trace over a fixed set of checkable predicates. It then specifies required behaviors using assume-guarantee contracts in linear temporal logic over finite traces (LTLf). Each contract is compiled to a deterministic finite automaton (DFA) that serves two roles: gating agent actions online and evaluating recorded traces offline. A contract library, acting as a reusable knowledge base, is maintained independently of the agent's model and can be applied across different agents within the same task domain. We show the effectiveness of our approach on four benchmarks spanning both roles, where ContrAgent matches state-of-the-art LLM-judge and rule-based guardrail baselines while producing deterministic, reproducible verdicts and, in the online mode, orders-of-magnitude lower per-call latency.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 19 of its 46 references that are in the map. First posted on arXiv 16 Sep 2026.

Builds on

19 of 46 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025SOPBench: Evaluating Language Agents at Following Standard Operating Procedures and Constraints11 Mar

Other references · 18

  1. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,393
  2. 2023SWE-bench: Can Language Models Resolve Real-World GitHub Issues?cited by 3,193
  3. 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 1,906
  4. 2023Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversationscited by 1,124
  5. 2024R-Judge: Benchmarking Safety Risk Awareness for LLM Agentscited by 268
  6. 2025AgentSpec: Customizable Runtime Enforcement for Safe and Reliable LLM Agentscited by 196
  7. 2025ShieldAgent: Shielding Agents via Verifiable Safety Policy Reasoningcited by 144
  8. 2025AgentAuditor: Human-Level Safety and Security Evaluation for LLM Agentscited by 82
  9. 2023Plug in the Safety Chip: Enforcing Constraints for LLM-driven Robot Agentscited by 67
  10. 2025RTBAS: Defending LLM Agents Against Prompt Injection and Privacy Leakagecited by 64
  11. 2023Grounding Complex Natural Language Commands for Temporal Tasks in Unseen Environmentscited by 52
  12. 2025VeriGuard: Enhancing LLM Agent Safety via Verified Code Generationcited by 46
6 more
  1. 2025Safer or Luckier? LLMs as Safety Evaluators Are Not Robust to Artifactscited by 25
  2. 2025ProbGuard: Proactive Runtime Monitoring for LLM Agent Safety via Probabilistic Predictioncited by 25
  3. 2026The Evolution of Tool Use in LLM Agents: From Single-Tool Call to Multi-Tool Orchestrationcited by 11
  4. 2022Monitoring Arithmetic Temporal Properties on Finite Tracescited by 7
  5. 2026Willful Disobedience: Automatically Detecting Failures in Agentic Tracescited by 6
  6. 2022Computing unsatisfiable cores for LTLf specificationscited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Contract2Tool: Learning Preconditions and Effects for Reliable Tool-Augmented LLM Agents
  2. 2025Empirical Characterization of Temporal Constraint Processing in LLMs
  3. 2026Formal Skill: Programmable Runtime Skills for Efficient and Accurate LLM Agents
  4. 2026From Agent Loops to Structured Graphs:A Scheduler-Theoretic Framework for LLM Agent Execution
  5. 2026ActGov: Governing LLM Agent Actions via Policy-Constrained Validation
  6. 2026LLM-as-Code: Agentic Programming for Agent Harness
  7. 2025An Approach to Checking Correctness for Agentic Systems
  8. 2025Enforcing Temporal Constraints for LLM Agents