FlagThis

Abstract

Background: Iterative feedback loops have become the dominant paradigm for improving LLMgenerated Infrastructure-as-Code (IaC): validators such as Checkov and `terraform validate` feed error signals back to the model for successive repair attempts. Prior work reports cumulative-best metrics, which are monotonically non-decreasing by construction, so the raw per-iteration security trajectory has never been examined in the IaC domain. Aims: We study security regression (a previously-passing CIS Benchmark check that fails after a repair iteration) to determine whether, and how often, iterative LLM repair degrades security while fixing other issues. Method: We analyze 5,968 scenario timelines from the IaC-Eval benchmark, each one scenario run through one configuration for up to 5 repair iterations. The 15 configurations comprise six model-specific RAG and nine model-aggregated non-RAG configurations, three temperatures each, and together they yield 4,440 iteration transitions with Checkov data on both sides. We track 30 individual CIS check IDs and classify regression root causes from code diffs, under two detection modes: standard (inclusive) and strict (exclusive check failures only). Results: Under standard (inclusive) detection, 13.8% of scenarios (24.8% of transitions) exhibit at least one regression. Under strict detection, which counts only unambiguous, exclusive check failures, the rate falls to 3.3% of scenarios (5.2% of transitions). This gap indicates that most apparent regressions are multi-resource measurement artifacts rather than genuine exclusive failures. Resource restructuring (79.0%) is the dominant root cause. Regression transitions show 2.6 __ more code churn (Cohens d = 0 . 90) and 4.9 __ higher strict-mode check volatility ( d = 1 . 49). Of standard-mode regressions, 36.6% self-correct within an average of 1.2 iterations, and iteration 3 is the optimal stopping point. Conclusions: Iterative IaC repair does introduce security regressions, but most apparent regressions are multi-resource measurement artifacts. The conservative, defensible rate is approximately 3.3% of scenarios. Our findings motivate security-aware feedback-loop design and provide actionable iteration-budget guidance.