Do Tool Calls Execute as Intended? Measuring and Repairing Intent-Execution Correspondence in LLM Agents
cs.AI, cs.SE
Submitted: 2026-10-03
Updated: 2026-10-03
Code: https://github.com/GLEAM-Lab/intact-bash-mod
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Entity Binding Failures in Tool-Augmented Agents
- $\tau^2$-Bench: Evaluating Conversational Agents in a Dual-Control Environment
- Chronicle: Cut-Point Replay for Regression Testing of LLM Agents
- Understanding Agent-Reactive Bugs at the Model-Harness Boundary: An Empirical Study of LLM Agent Issue Reports
- TUA-Bench: A Benchmark for General-Purpose Terminal-Use Agents
- Get Experience from Practice: LLM Agents with Record & Replay
- Dissecting model behavior through agent trajectories
- MAS-FIRE: Fault Injection and Reliability Evaluation for LLM-Based Multi-Agent Systems
- DFAH-Bench: Benchmarking Observable Agent Instability in Financial Decision-Making
- QuoteBench: How Matched Scores Can Hide Command-Path Failures
- Dive into Claude Code: The Design Space of Today's and Future AI Agent Systems
- An Empirical Study on Failures in Automated Issue Solving
- MCP-Universe: Benchmarking Large Language Models with Real-World Model Context Protocol Servers
- Understanding Code Agent Behaviour: An Empirical Study of Success and Failure Trajectories
- Securing LLM Agents Need Intent-to-Execution Integrity
- Context-to-Execution Integrity for LLM Agents
- Causal Agent Replay: Counterfactual Attribution for LLM-Agent Failures
- AgentChaos: Chaos Engineering for Agent Systems via Programmatic Fault Injection
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems