jira

In project management, existing tools impose high learning costs and require programming expertise for data retrieval, hindering real-time decision-making. This paper introduces a lightweight, natural-language-interfacing plugin for Jira, built upon the GPT large language model (LLM). By integrating prompt engineering with the Jira REST API, the plugin enables non-programmatic task querying and status management. Its key innovation lies in natively embedding an LLM directly into the project management workflow and systematically evaluating diverse prompt strategies—such as zero-shot, few-shot, and chain-of-thought—against information retrieval accuracy and response latency. Experimental results demonstrate that optimized prompting improves critical task identification accuracy by 27.4% and reduces average response time by 39%, significantly lowering cognitive load. The study validates the practical utility and deployment feasibility of LLMs in domain-specific human-AI collaboration.

This study systematically evaluates the applicability of mainstream large language models (LLMs)—namely GPT-3.5 Turbo, GPT-4 Turbo, and Val—to Scrum iteration planning, focusing on three core tasks: user story estimation, task decomposition, and sprint goal generation. Using a manually curated, real-world annotated dataset, we employ a mixed qualitative and quantitative evaluation to empirically assess LLM outputs across accuracy, consistency, and operational feasibility—the first such investigation into engineering-grade usability in agile planning. Results indicate that current LLM outputs do not yet meet the threshold for direct production deployment. Key contributions include: (1) an empirical delineation of LLM capabilities and limitations in Scrum planning; (2) a practical hybrid enhancement framework combining rule-based engines with lightweight fine-tuning; and (3) the first empirically grounded benchmark and improvement roadmap for LLMs in Scrum iteration planning—advancing the integration of foundation models into software engineering practice.

To address inefficient sprint planning and perfunctory retrospective meetings in Agile/Scrum, this paper proposes RetroAI++—the first lightweight temporal reasoning framework integrating meeting transcripts, task logs, and code-level behavioral traces. Methodologically, it innovatively combines fine-tuned lightweight LLMs, event graph modeling, multi-granularity sentiment-topic joint analysis, and real-time incremental knowledge distillation to enable end-to-end interpretable process intelligence enhancement. Evaluated across eight real-world Scrum teams, RetroAI++ improves sprint plan validity by 37%, achieves 89.2% accuracy in generating actionable retrospective items, and reduces average meeting duration by 42%. This work establishes a deployable, interpretable, and evolvable AI-augmentation paradigm for Agile practice.

To address the high cost and low efficiency of manual triage for bug reports in large-scale software projects, this paper proposes a GitHub-integrated AI assistant that, for the first time, unifies multi-task deep learning across the full bug-report management pipeline—including duplicate detection, severity prediction, and fix-file recommendation. The method integrates a BERT-based semantic model, a graph neural network (GNN) for code localization, and a lightweight GitHub App architecture to deliver end-to-end interpretable recommendations. Evaluated on multiple benchmark datasets, the approach achieves F1 scores of 0.82–0.89. A real-world user study with professional developers demonstrates an average 47% reduction in triage time and a 76% recommendation adoption rate. These results significantly advance the automation level and practical utility of defect management in industrial settings.

This work addresses the lack of closed-loop control in traditional software development lifecycles, which often fails to simultaneously ensure security, auditability, and highly reliable automation. The authors propose a deterministic autonomous control framework that models the lifecycle as a seven-stage automated pipeline, integrating Jira-based task orchestration, structured context, resource constraints, and human-review gating mechanisms to establish a secure closed loop. Key innovations include a state-contract-based collision locking mechanism, a degradation protocol for fallback operation, and a traceable control architecture. Implemented with 12,661 lines of Python code and 6,907 lines of versioned prompt specifications—including 101 exception handlers and 12 centralized locks—the system achieved a 100% success rate (95% CI [97.6%, 100%]) across 152 initial runs, producing over 795 artifacts. All 51 issues identified through adversarial review were fully resolved, with 60% of security tickets autonomously completed.

This study addresses the significant burden developers face in authoring and maintaining GitHub Actions workflows, stemming from a lack of systematic understanding of real-world automation and reuse practices. Through a mixed-methods approach combining a survey of 419 practitioners with qualitative and quantitative analysis, this work presents the first developer-centric characterization of common automation tasks, patterns of reuse mechanism adoption, and maintenance pain points in workflow development. The findings reveal that while developers heavily rely on reusable Actions, they seldom adopt reusable workflows; version management challenges lead to rampant copy-pasting; and critical aspects such as security and performance monitoring remain under-automated. These insights provide empirical foundations for improving CI/CD toolchains and reuse mechanisms.

Traditional defect tracking relies on manual reporting, reproduction, and classification, resulting in inefficient cross-role communication and delayed responses. Method: This paper proposes the first large language model (LLM)-integrated intelligent defect tracking framework, enabling end-to-end automation—from natural-language user reports to defect localization, automatic root-cause attribution, category prediction, and candidate patch generation. The framework combines AI agents with advanced NLP techniques, supports no-code remediation responses, and is embeddable into existing SaaS platforms. Contribution/Results: Experiments demonstrate significant reductions in defect response and resolution cycles, a 72% decrease in manual intervention, and improved cross-role collaboration efficiency and user satisfaction. This work provides the first systematic empirical validation of LLM-driven full-lifecycle automation in software maintenance, confirming both its feasibility and effectiveness.

This study addresses the lack of systematic understanding regarding how GitHub Actions workflows are used in real-world scenarios, how developers respond to workflow failures, and how these practices relate to project characteristics. Combining large-scale quantitative analysis of 258,300 workflow runs with qualitative case studies across 21 diverse repositories, this work identifies three typical patterns developers employ to handle workflow failures and uncovers a “configuration–usage gap”—where YAML configurations exist but workflows remain effectively unused. Furthermore, the study empirically validates five hypotheses linking project features to workflow usage intensity, revealing a significant positive correlation between high usage intensity and low failure rates. These findings provide actionable empirical evidence for improving CI/CD practices.

This study addresses the lack of effective cross-channel traceability between community forums and issue trackers in open-source software, which hinders collaboration and transparency throughout the feature request lifecycle. Focusing on the Moodle platform, the research integrates empirical analysis of forum and Jira data, semi-structured interviews, and case studies to systematically reveal that cross-channel links are highly sparse—only approximately 3.5% of issues reference forum discussions—and that author roles are distinctly differentiated, with users predominantly initiating forum posts while developers primarily create tracker issues. Furthermore, the conversion process from forum requests to tracked issues is characterized by informality and low responsiveness. The findings underscore critical gaps in tooling support and ambiguous accountability, offering empirical evidence and design implications for improving collaborative infrastructure in open-source ecosystems.

Existing research lacks an open, real-world Jira–oriented text-to-JQL evaluation benchmark. This paper introduces Jackal—the first large-scale, execution-based JQL generation benchmark grounded in authentic Jira usage, comprising 100K natural language query–executable JQL pairs spanning diverse user intent categories. Methodologically, we propose a novel multi-dimensional evaluation framework grounded in dynamic execution against live Jira instances, measuring exact match, canonical match, and execution accuracy. We publicly release a reproducible Jira snapshot and an open-source scoring toolkit. Evaluating 23 large language models on the Jackal-5K subset, Gemini 2.5 Pro achieves the highest execution accuracy (60.3%), yet exhibits substantial performance variance across query types—revealing fundamental limitations in short-text comprehension and semantic similarity reasoning.