• Broaden the sampling or narrow the claims of representativeness. The bacterial collection was derived from only 12 floor-laid eggs obtained from four backyard keepers, rather than from commercial production systems. Although 50 isolates were identified and eight species were selected for detailed analysis, this sampling frame is insufficient to support broad conclusions about eggshell-associated microbial diversity or commercial egg processing. The authors should ideally include eggs from multiple commercial flocks or production batches. If additional sampling is not feasible, the study should be clearly presented as preliminary and the conclusions restricted to the isolates and backyard eggs examined.

• Clarify the detection limit and avoid equating loss of culturability with complete killing. PAW efficacy was assessed exclusively by CFU enumeration, yet the precise limit of detection is not reported. Statements such as “complete inactivation” should therefore be replaced with wording such as “no culturable cells were recovered above the assay detection limit.” The authors should report the exact detection limit and acknowledge that culture-based enumeration cannot exclude viable-but-non-culturable cells. An independent viability measurement, such as membrane-integrity staining combined with microscopy or flow cytometry, would strengthen the bactericidal claim.

• Validate PAW directly on eggshells. Despite the proposed application, the PAW experiments used biofilms formed on plastic microtitre plates, with SEM experiments conducted on plastic coverslips. The authors should test PAW against biofilms established directly on eggshell fragments or intact eggs, preferably incorporating realistic organic contamination and an exposure time relevant to commercial washing, which the manuscript describes as approximately 30 seconds. A comparison with an established eggshell sanitiser would further improve the practical relevance. If these experiments cannot be added, claims regarding application in egg production should be substantially moderated.

• Improve structure and separation of sections: Reorganize the Results into clear subheadings and restrict it to direct experimental observations. Move mechanistic interpretation, literature context, and clinical implications to the Discussion. New findings belong in Results, but new background concepts should be introduced earlier or discussed later.

• Correct all in-text references and presentation details: Align figure and table citations with the actual numbering (e.g., Figures 1–7 and Table 1) and standardize supplementary references, units, labels, and terminology throughout. Figure legends should describe the experiment and data without interpreting the mechanism.

• Strengthen the Discussion and future-work section: Avoid repeating the Results. Instead, discuss the significance, limitations, alternative explanations, and strength of the mechanistic evidence. Future work should validate the heme and beta-lactamase mechanisms directly, test clinical isolates and wound-like models, and compare turbidity-based MICs with CFU, time-kill, and bactericidal endpoints.

• Appropriate carrier controls and lipid-dose reporting Include an empty-LNP control, prepared and purified using the same procedure, at lipid doses matched to each AuNCs@LNP group in the principal in vitro and in vivo experiments. The final lipid concentration, injected lipid mass, and animal exposure in mg/kg should also be reported.

• Statistical analysis and reporting Strengthen the statistical reporting by providing the prespecified direct comparison between the 75 and 150 µg AuNCs@LNP groups, together with exact adjusted p-values, effect sizes, and 95% confidence intervals. The meaning of all significance symbols should be defined. The discrepancy between the Dunnett tests stated in the Figure 2 and Figure 3 legends and the Bonferroni procedure reported in the Methods should also be resolved.

• Consistency of Methods, figure legends, and assay controls Systematically reconcile the Methods and figure legends regarding doses, replicate numbers, controls, and statistical procedures. In particular, appropriate background controls should be included for the ROS and XTT assays to exclude optical interference. The normalization procedure and the use of separate parallel wells should also be clearly described and justified.

• Statistical analysis and reporting: Distinguish biological from technical replicates, present the results from all independent experiments, describe the statistical methods and their limitations, and avoid claims of statistical significance where a test could not be performed. Exact p-values or appropriate measures of variability should be provided where possible.

• Interpretation of mechanism: The current data support an interaction-associated change in the C. jejuni phenotype but do not establish the underlying molecular mechanism. Proposed roles for direct contact, microcolony formation, altered protein expression, or the cmeABC efflux system should either be experimentally investigated or clearly identified as hypotheses.

• Controls and causal validation: Please provide full information on assay controls and replication. Reconstruction of the matched purified C. jejuni–Enterococcus pairs at controlled starting ratios, followed by testing against multiple relevant antibiotics, would substantially strengthen the conclusion that the associated enterococci are sufficient to produce the observed phenotype.