International Journal of Plant & Soil Science

Vegetable crops occupy a small share of global cropland yet supply a disproportionate fraction of dietary micronutrients, and their short life cycles, high water demand and thermosensitive reproductive stages make them unusually exposed to a warming and more variable climate. Clustered regularly interspaced short palindromic repeats and CRISPR-associated protein 9 (CRISPR-Cas9) editing has been promoted as a route to climate-resilient vegetable cultivars, and the descriptive literature cataloguing edited targets has expanded quickly. This critical narrative review evaluates whether that literature supports the resilience claims made for it. Peer-reviewed studies published between January 2015 and June 2026 were identified through openly accessible bibliographic indexes and citation-metadata services, appraised for design adequacy, evaluation environment and evidence-claim alignment, and synthesised thematically rather than catalogued. Three findings dominate. First, the evidence base is heavily weighted towards single-gene loss-of-function alleles created in a small number of transformation-competent model genotypes, assessed under controlled conditions using physiological proxies rather than marketable yield, so that demonstrated gene function is repeatedly reported in language implying demonstrated agronomic improvement. Second, several loci yield directionally opposite phenotypes depending on the stress applied, which indicates that stress-specific reactive oxygen species thresholds and pleiotropy are poorly resolved and that resilience cannot be treated as a single transferable trait. Third, the small number of studies that combine allele-resolution editing of regulatory sequence with field evaluation and yield endpoints produce markedly stronger inference than the knockout literature, suggesting that the principal constraint is experimental design and delivery capability rather than nuclease performance. Regulatory divergence, patent uncertainty and uneven transformation capacity across vegetable species further separate laboratory demonstration from field deployment. Confidence in current claims of climate resilience remains low for most targets, and progress will depend on multi-environment field testing, yield-based endpoints, and extension of editing capability beyond the few genotypes that presently regenerate reliably.