# Testing and Understanding Erroneous Planning in LLM Agents through Synthesized User Inputs

arXiv: https://arxiv.org/abs/2404.17833 · HTML: https://mlsanity.com/paper/2404.17833/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- Builds on 42 of its 62 references that are in the map.
- First posted on arXiv 27 Apr 2024.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 30 of 42)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Do As I Can, Not As I Say: Grounding Language in Robotic Affordances](https://mlsanity.com/paper/2204.01691/) (2022)
- [Large Language Models Encode Clinical Knowledge](https://mlsanity.com/paper/2212.13138/) (2022)
- [Jailbroken: How Does LLM Safety Training Fail?](https://mlsanity.com/paper/2307.02483/) (2023)
- [ALFWorld: Aligning Text and Embodied Environments for Interactive Learning](https://mlsanity.com/paper/2010.03768/) (2020)
- [HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face](https://mlsanity.com/paper/2303.17580/) (2023)
- [Large Language Models Cannot Self-Correct Reasoning Yet](https://mlsanity.com/paper/2310.01798/) (2023)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [InjecAgent: Benchmarking Indirect Prompt Injections in Tool-Integrated Large Language Model Agents](https://mlsanity.com/paper/2403.02691/) (2024)
- [LLM+P: Empowering Large Language Models with Optimal Planning Proficiency](https://mlsanity.com/paper/2304.11477/) (2023)
- [PlanBench: An Extensible Benchmark for Evaluating Large Language Models on Planning and Reasoning about Change](https://mlsanity.com/paper/2206.10498/) (2022)
- [Causal Reasoning and Large Language Models: Opening a New Frontier for Causality](https://mlsanity.com/paper/2305.00050/) (2023)
- [Character-LLM: A Trainable Agent for Role-Playing](https://mlsanity.com/paper/2310.10158/) (2023)
- [Selection-Inference: Exploiting Large Language Models for Interpretable Logical Reasoning](https://mlsanity.com/paper/2205.09712/) (2022)
- [Hallucination is Inevitable: An Innate Limitation of Large Language Models](https://mlsanity.com/paper/2401.11817/) (2024)
- [On the Planning Abilities of Large Language Models : A Critical Investigation](https://mlsanity.com/paper/2305.15771/) (2023)
- [Evaluating Large Language Models at Evaluating Instruction Following](https://mlsanity.com/paper/2310.07641/) (2023)
- [R-Judge: Benchmarking Safety Risk Awareness for LLM Agents](https://mlsanity.com/paper/2401.10019/) (2024)
- [MetaTool Benchmark for Large Language Models: Deciding Whether to Use Tools and Which to Use](https://mlsanity.com/paper/2310.03128/) (2023)
- [Is ChatGPT A Good Translator? Yes With GPT-4 As The Engine](https://mlsanity.com/paper/2301.08745/) (2023)
- [SmartPlay: A Benchmark for LLMs as Intelligent Agents](https://mlsanity.com/paper/2310.01557/) (2023)
- [SatLM: Satisfiability-Aided Language Models Using Declarative Prompting](https://mlsanity.com/paper/2305.09656/) (2023)
- [CriticBench: Benchmarking LLMs for Critique-Correct Reasoning](https://mlsanity.com/paper/2402.14809/) (2024)
- [Can large language models build causal graphs?](https://mlsanity.com/paper/2303.05279/) (2023)
- [Split and Merge: Aligning Position Biases in LLM-based Evaluators](https://mlsanity.com/paper/2310.01432/) (2023)
- [A Causal Framework to Quantify the Robustness of Mathematical Reasoning with Language Models](https://mlsanity.com/paper/2210.12023/) (2022)
- [GPTScan: Detecting Logic Vulnerabilities in Smart Contracts by Combining GPT with Program Analysis](https://mlsanity.com/paper/2308.03314/) (2023)
- [LLM4Vuln: A Unified Evaluation Framework for Decoupling and Enhancing LLMs' Vulnerability Reasoning](https://mlsanity.com/paper/2401.16185/) (2024)

## Built on by (most cited first, 22 of 22)

- [Multi-Agent Collaboration Mechanisms: A Survey of LLMs](https://mlsanity.com/paper/2501.06322/) (2025)
- [Where LLM Agents Fail and How They can Learn From Failures](https://mlsanity.com/paper/2509.25370/) (2025)
- [AI Agents Under Threat: A Survey of Key Security Challenges and Future Pathways](https://mlsanity.com/paper/2406.02630/) (2024)
- [OctoTools: An Agentic Framework with Extensible Tools for Complex Reasoning](https://mlsanity.com/paper/2502.11271/) (2025)
- [ACE: A Security Architecture for LLM-Integrated App Systems](https://mlsanity.com/paper/2504.20984/) (2025)
- [VeriPlan: Integrating Formal Verification and LLMs into End-User Planning](https://mlsanity.com/paper/2502.17898/) (2025)
- [SAMULE: Self-Learning Agents Enhanced by Multi-level Reflection](https://mlsanity.com/paper/2509.20562/) (2025)
- [AgentForesight: Online Auditing for Early Failure Prediction in Multi-Agent Systems](https://mlsanity.com/paper/2605.08715/) (2026)
- [TAI3: Testing Agent Integrity in Interpreting User Intent](https://mlsanity.com/paper/2506.07524/) (2025)
- [When Actions Go Off-Task: Detecting and Correcting Misaligned Actions in Computer-Use Agents](https://mlsanity.com/paper/2602.08995/) (2026)
- [A Good Plan is Hard to Find: Aligning Models with Preferences is Misaligned with What Helps Users](https://mlsanity.com/paper/2509.18632/) (2025)
- [RAG or Fine-tuning? A Comparative Study on LCMs-based Code Completion in Industry](https://mlsanity.com/paper/2505.15179/) (2025)
- [Mobile-Agent-RAG: Driving Smart Multi-Agent Coordination with Contextual Knowledge Empowerment for Long-Horizon Mobile Automation](https://mlsanity.com/paper/2511.12254/) (2025)
- [Automated Penetration Testing with LLM Agents and Classical Planning](https://mlsanity.com/paper/2512.11143/) (2025)
- [Teaching AI to Handle Exceptions: Supervised Fine-Tuning with Human-Aligned Judgment](https://mlsanity.com/paper/2503.02976/) (2025)
- [PIVOT: Bridging Planning and Execution in LLM Agents via Trajectory Refinement](https://mlsanity.com/paper/2605.11225/) (2026)
- [Agent WARPP: Workflow Adherence via Runtime Parallel Personalization](https://mlsanity.com/paper/2507.19543/) (2025)
- [TREAT: A Code LLMs Trustworthiness / Reliability Evaluation and Testing Framework](https://mlsanity.com/paper/2510.17163/) (2025)
- [Assessing Deanonymization Risks with Stylometry-Assisted LLM Agent](https://mlsanity.com/paper/2602.23079/) (2026)
- [Mosaic: Runtime-Efficient Multi-Agent Embodied Planning](https://mlsanity.com/paper/2607.09603/) (2026)
- [HPFA: Hypergraph-Based Paired Failure Attribution for LLM Reasoning](https://mlsanity.com/paper/2608.02026/) (2026)
- [On the Suitability of LLM-Driven Agents for Dark Pattern Audits](https://mlsanity.com/paper/2603.03881/) (2026)

## Abstract (arXiv)

Agents based on large language models (LLMs) have demonstrated effectiveness in solving a wide range of tasks by integrating LLMs with key modules such as planning, memory, and tool usage. Increasingly, customers are adopting LLM agents across a variety of commercial applications critical to reliability, including support for mental well-being, chemical synthesis, and software development. Nevertheless, our observations and daily use of LLM agents indicate that they are prone to making erroneous plans, especially when the tasks are complex and require long-term planning. In this paper, we propose PDoctor, a novel and automated approach to testing LLM agents and understanding their erroneous planning. As the first work in this direction, we formulate the detection of erroneous planning as a constraint satisfiability problem: an LLM agent's plan is considered erroneous if its execution violates the constraints derived from the user inputs. To this end, PDoctor first defines a domain-specific language (DSL) for user queries and synthesizes varying inputs with the assistance of the Z3 constraint solver. These synthesized inputs are natural language paragraphs that specify the requirements for completing a series of tasks. Then, PDoctor derives constraints from these requirements to form a testing oracle. We evaluate PDoctor with three mainstream agent frameworks and two powerful LLMs (GPT-3.5 and GPT-4). The results show that PDoctor can effectively detect diverse errors in agent planning and provide insights and error characteristics that are valuable to both agent developers and users. We conclude by discussing potential alternative designs and directions to extend PDoctor.