n8n

This study addresses the inefficiency and error-proneness of business processes in small-scale enterprises stemming from limited in-house development expertise. To mitigate this, the authors designed and deployed a representative lead-processing workflow on the n8n low-code automation platform, integrating automated data storage, email confirmation, and real-time notifications. A controlled experiment conducted in a real-world business setting was employed to quantitatively evaluate the performance of n8n-based automation against manual execution—the first such empirical assessment reported in the literature. The results demonstrate that automation reduced the average execution time from 185.35 seconds to 1.23 seconds (a ~151-fold speedup) and eliminated errors entirely, decreasing the error rate from 5% to 0%. These findings underscore the substantial gains in both efficiency and reliability achievable through low-code tools, affirming their practical utility and deployability for non-technical users.

This study addresses the lack of systematic understanding regarding the practical usage patterns, reliability mechanisms, and autonomy levels of large language model (LLM) agents in low-code/no-code platforms. Drawing on over 6,000 publicly available n8n workflows, the authors employ large-scale data mining, structured log analysis, and qualitative coding to empirically characterize how LLM agents are deployed in real-world automation scenarios—specifically examining task distribution, workflow structure, tool invocation, and degrees of autonomy. The findings reveal that while LLMs are commonly embedded within complex workflows featuring control logic and human review steps, such workflows generally lack structured fault tolerance, repair loops, and approval mechanisms. Based on these insights, the study articulates ten empirical observations and five design implications to inform the development of more reliable and governable low-code platforms.

Existing browser automation tools suffer from insufficient environmental fidelity, poor reproducibility, weak robustness, and high operational costs when generating diverse, high-fidelity network traffic. To address these limitations, we propose a natural language–driven executable workflow compilation framework. It automatically compiles user-specified, state-dependent web application behaviors—expressed in natural language—into nondeterministic finite automata (NFAs). By integrating state caching and LLM invocation optimization, the framework ensures execution determinism while enhancing adaptability to dynamic UI variations. The approach enables rapid workflow reconstruction and precise trajectory replay, substantially reducing manual intervention and computational overhead. Evaluated across 50+ real-world scenarios—including video-on-demand, live streaming, online conferencing, social media interaction, and web crawling—the framework generates high-fidelity traffic traces, demonstrating superior performance in diversity, robustness, and efficiency.

This work demonstrates that large language model (LLM) agents integrated into automation platforms are vulnerable to manipulation via malicious inputs—such as GitHub comments—leading to risks including credential leakage and arbitrary command execution. To systematically uncover these novel attack surfaces in agent workflows, the authors propose “Context-Anchored Evolution,” a method that combines static path feasibility analysis, dynamic prompt provenance tracing, and runtime capability assessment within a three-stage contextual framework, enhanced by evolutionary input generation to achieve targeted hijacking of LLM agents. The resulting JAW framework successfully exploits 4,714 GitHub workflows and 8 n8n templates, compromising 15 widely used GitHub Actions and 2 official n8n nodes. Vulnerabilities identified through this approach have been acknowledged and patched by vendors including GitHub, Google, and Anthropic.

Existing LLM-based agent workflows rely on manually designed instructions and action sequences, suffering from poor scalability. This paper proposes the first end-to-end fully automated workflow generation framework, formulating workflow optimization as a tree search problem in code representation space. Our method innovatively integrates Monte Carlo Tree Search (MCTS), LLM-driven node orchestration, and execution-feedback-guided code-level iterative refinement—requiring no human initialization. It achieves complete automation from task input to executable workflow generation and optimization. Evaluated on six benchmark tasks, our approach improves average accuracy by 5.7% over prior methods. Notably, a lightweight model achieves performance superior to GPT-4o while incurring only 4.55% of its inference cost.

This work addresses the limitation of existing distributed data pipeline systems, which require users to explicitly define complete workflow graphs, by proposing a unified planning and scheduling framework that automatically constructs end-to-end persistent pipelines from implicit goal declarations alone. The approach introduces, for the first time, a numeric-domain-independent planner into the context of persistent scheduling, integrating workflow and resource graph modeling, numeric planning, and network interface scheduling to achieve full automation. Experimental results demonstrate the feasibility and scalability of the method: under a single-machine constraint of one hour of CPU time and 30 GB of memory, the system successfully scheduled a linear pipeline spanning eight sites and comprising fourteen components.

Industrial-scale executable visual workflows are typically handcrafted, resulting in high costs, error-proneness, and limited adaptability to evolving requirements. This work proposes a framework based on large language models and an agent architecture that enables automatic generation of deployable visual workflows—from natural language specifications—for platforms such as Dify and Coze. The study introduces the first benchmark dataset for natural language-to-executable workflow generation grounded in real-world business scenarios and incorporates an agent mechanism specifically designed to mitigate execution errors. Experimental results demonstrate that existing models still struggle to reliably generate correct workflows for complex tasks, whereas the proposed approach improves the execution error resolution rate by up to 5.34%, establishing a new foundation for industrial automation.

Enterprise operational workflows are notoriously difficult to automate end-to-end due to their heavy reliance on human intervention and limited adaptability to change. This work proposes the first action-centric workflow graph framework, which achieves automated construction, execution, and evolution through a three-stage pipeline: structured workflow graphs are extracted from human operation traces, executed via multi-agent online traversal, and continuously optimized in a closed loop using an Adaptive Traversal Reinforcement (ATR) mechanism. Integrating large-scale offline graph construction, graph-guided retrieval, and large language model reasoning, the approach was deployed across four cloud database services. It substantially outperforms the Trace-RAG baseline in coverage breadth, factual accuracy, and diagnostic throughput, achieving an expert blind-review score of 4.95 out of 5.

Current evaluations of AI agents lack comprehensive assessment of cross-application coordination, autonomous API discovery, and policy adherence, failing to reflect the demands of real-world business automation. This work proposes the first unified benchmark that integrates cross-application workflow orchestration, autonomous API exploration, and regulatory compliance. Built upon the Zapier platform, the benchmark encompasses multi-scenario tasks spanning CRM, email, calendar, and other systems, requiring agents to autonomously discover APIs, follow multi-layered business rules, and execute cross-system data writes in realistic sales and marketing workflows. Evaluation relies solely on end-state correctness through an automated scoring mechanism. Experimental results reveal that even state-of-the-art models achieve success rates below 10%, highlighting a significant capability gap in deploying current AI systems for practical business automation.

This study addresses the challenge of automating workflows in complex industries—such as logistics, healthcare, and construction—where processes are fragmented across heterogeneous tools and involve multi-party collaboration. The work proposes orchestration as a core abstraction to enable effective automation by dynamically coordinating multi-step tasks, enforcing domain-specific constraints, managing human approvals, and integrating legacy systems. It introduces the novel concept of “orchestration bottlenecks” and develops a theoretical framework that unifies multi-agent systems, workflow modeling, constraint reasoning, and human–AI collaboration, while exposing critical gaps in current multi-agent approaches at the orchestration level. Based on distinct sources of operational friction across domains, the paper advocates for targeted architectural safeguards—such as constraint enforcement or explainability—and phased implementation strategies to provide actionable pathways for automation in complex operational environments.