EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An Empirical Study of Security Smells in Python Applications: Identification, Impact, and Mitigation Practices

Reem A. Almutairi, Noureddine Abbadeni

IEEE Access · 2026 · 14 · 62524-62546

DOI: 10.1109/access.2026.3686254 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Reem A. Almutairi und 1 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Security Vulnerabilities/Security Smells (SVS) are recurring code or configuration patterns that indicate potential or latent security weaknesses in software systems. Despite growing interest in this area, limited empirical evidence exists on how such issues are recognized, perceived, and mitigated in Python applications across Software Development Life Cycle (SDLC) phases. This study investigates how Python developers and security professionals identify, perceive, and mitigate SVS through a structured practitioner survey. A total of 156 responses were collected, of which 104 valid responses were retained after eligibility screening. The survey examines recognition, perceived frequency, impact, and mitigation practices for 22 predefined SVS mapped to the Open Web Application Security Project (OWASP) Top 10 (2021) and MITRE Common Weakness Enumeration (CWE) standards. SQL Injection was the most recognized vulnerability (62.5%) and ranked second in perceived impact after Remote Code Execution, followed by Command Injection (48.1% recognition), while Remote Code Execution, despite lower reported frequency, received a 60.6% high-impact rating. Vulnerability detection was primarily concentrated in the testing phase (70.2%), whereas only 42.3% of respondents reported systematic security integration across all SDLC stages. Common mitigation practices included penetration testing (66.3%), regular code reviews (61.5%), and automated security scanning (55.8%), although proactive techniques such as threat modeling (20.2%) remain limited. Spearman’s rank correlation showed no significant association between perceived frequency and impact ( $\rho =0.126$ , p = 0.577), suggesting frequency-based prioritization may overlook high-impact vulnerabilities. These findings provide practitioner-centered evidence supporting risk-aware prioritization and stronger integration of proactive security practices in Python development workflows.

Zitieren

APA

Almutairi, R. A., & Abbadeni, N. (2026). An Empirical Study of Security Smells in Python Applications: Identification, Impact, and Mitigation Practices. IEEE Access, 14, 62524-62546. https://doi.org/10.1109/access.2026.3686254

Vancouver

Almutairi RA, Abbadeni N. An Empirical Study of Security Smells in Python Applications: Identification, Impact, and Mitigation Practices. IEEE Access. 2026;14:62524-62546. doi:10.1109/access.2026.3686254

BibTeX
@article{Almutairi2026,
  author = {Almutairi, Reem A. and Abbadeni, Noureddine},
  title = {An Empirical Study of Security Smells in Python Applications: Identification, Impact, and Mitigation Practices},
  journal = {IEEE Access},
  year = {2026},
  volume = {14},
  pages = {62524--62546},
  doi = {10.1109/access.2026.3686254}
}
Kostenlos starten →