cs.SENov 11, 2026

GHA-SmellFix: Taking Action with Agentic LLMs for Detecting and Refactoring GitHub Actions Workflow Smells

Authors: Anderson Uchôa, Jhefferson Souza, Emanuel Ávila, Gabriel Silva, Kevin Lira, Hui Sun, Nelson Araújo, Arthur Cordeiro, 

Organizations: Federal University of Ceará, Brazil · North Carolina State University, USA · Federal University of Campina Grande, Brazil

Abstract

Continuous integration and continuous delivery workflows are key artifacts in software projects. In GitHub Actions, they encode build, test, release, deployment, and security automation in YAML. These workflows contain configuration smells affecting security, reliability, maintainability, or operational stability. Existing approaches are mostly rule-based, with rules hard to create and evolve, and typically report findings without fixes. Agentic LLM pipelines can combine contextual detection, transformation selection, and patch generation. Specialized agents can gate refactoring on positive detection verdicts, potentially avoiding unnecessary changes, but it is unclear whether decomposition improves effectiveness or mainly changes overhead. We present GHA-SmellFix, an agentic LLM pipeline for detecting and refactoring GitHub Actions workflow smells. We compare two designs: Joint, where one agent detects and refactors, and Decomposed, where separate agents detect and then refactor. We operationalize 19 smells from an existing catalog with explicit detection criteria, refactoring expectations, exclusion rules, and safety checks, and curate a ground truth of 1,710 (workflow, smell) pairs. We evaluate the designs with four LLMs (GPT-5.4-nano, Gemini 3.5 Flash, Claude Haiku 4.5, and DeepSeek V4 Flash) on 570 sampled pairs. Results show that decomposition does not universally improve patch quality; its main benefit is controlling when patch generation is attempted. It helps models prone to missing smells and can reduce overhead by gating patches. Once a smell is correctly detected, clean refactoring is generally effective, making detection the main end-to-end bottleneck. However, false positives can trigger unnecessary edits to execution-relevant constructs. Practical CI/CD agents thus need reliable abstention and construct-aware validation alongside patch generation.

Figures & tables

Explore similar work

CardsList
  1. SmellBench: Evaluating LLM Agents on Architectural Code Smell Repair

    May 7, 2026Ion George Dinu, Marian Cristian Mihăescu, Traian RebedeaAutomated Program RepairLLM Agent Evaluation

  2. REFINE: A Multi-Agent LLM Approach for Evidence-Guided Code Refactoring

    Aug 21, 2026Muhammad Waseem, Aakash Ahmad, Pekka AbrahamssonAutomated Code RefactoringAutomated Software Engineering

  3. LLM-Driven CI-CD Workflow Intelligence for Cyber Systems Engineering

    Jul 6, 2026Bonan Shen, Jiazhou Gao, Tao Ning +2Process Mining