TerraProbe shows that targeted Checkov removal overstates LLM Terraform repair success, with 71.4% of plan-compared real-world repairs being deceptive fixes that leave vulnerabilities intact.
Aldegheri et al., ”GenSIaC: Generative Security IaC with LLMs,” arXiv:2511.12385, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Empirical Software Engineering TerraProbe: A Layered-Oracle Framework for Detecting Deceptive Fixes in LLM-Assisted Terraform
TerraProbe shows that targeted Checkov removal overstates LLM Terraform repair success, with 71.4% of plan-compared real-world repairs being deceptive fixes that leave vulnerabilities intact.