Phthalate esters (PAEs) are used as additives to enhance the pliability and malleability of plastics. These substances frequently migrate from packaging materials to vegetable oils because of the absence of covalent bonds. Over time, this migration could result in the accumulation of PAEs in the human body through ingestion, contributing to various diseases. Therefore, accurate qualitative and quantitative analyses of PAEs in vegetable oils are imperative to assess the origins of contamination and investigate their toxicity, degradation, migration, and transformation patterns. However, the concentration of PAEs in most samples is low, and the composition of vegetable oils is complex. Thus, PAEs must be enriched and purified using appropriate sample pretreatment procedures before analysis. Common methods for pretreating PAEs in oil include solid-phase extraction (SPE), dispersive SPE, and magnetic SPE. These techniques require time-consuming and labor-intensive procedures such as oil dissolution, solvent extraction, and degreasing. These approaches also require numerous solvents and containers, increasing the risk of sample cross-contamination. Solid-phase microextraction (SPME) integrates sampling, extraction, purification, concentration, and injection into a single process, significantly accelerating analytical testing and reducing the potential for sample cross-contamination. In headspace (HS) mode, the analytes achieve equilibrium on the coating and are extracted in the gas phase. The fibers are shielded from nonvolatile and high-relative molecular mass substances in the sample matrix. Thus, SPME is an ideal method for extracting volatile compounds in vegetable oils. When HS-SPME coupled with gas chromatography-mass spectrometry (GC-MS), it can achieve the rapid screening of PAEs in vegetable oil. In this study, an SPME with cyclodextrin-based hypercrosslinked polymers (BnCD-HCP) coated on stainless steel fibers was employed to extract PAEs from vegetable oil. The structure and morphology of the polymers were characterized using Fourier-transform infrared spectroscopy, nuclear magnetic spectroscopy, and scanning electron microscopy. BnCD-HCP exhibited high stability and diverse interactions, including π-π, hydrophobic, and host-guest interactions. The oil samples were incubated with methanol, and the PAEs were extracted from the headspace using the probe. The optimal extraction parameters included an extraction time of 20 min, extraction temperature of 50 ℃, desorption time of 4 min, and desorption temperature of 275 ℃. The BnCD-HCP/HS-SPME method was evaluated under optimized experimental conditions. The limits of detection (LODs) and quantification (LOQs) were determined by applying signal-to-noise ratios (S/N) of 3 and 10, respectively. Method accuracy was evaluated using relative standard deviations (RSDs). Single-needle precision was evaluated by conducting three consecutive analyses at 3 h intervals within a day. Inter-needle precision was assessed by conducting the same analyses (three replicates) with differently coated fibers. The 12 PAE compounds exhibited good linearity with correlation coefficients (R2) of at least 0.99. The LODs and LOQs ranged from 0.21 to 3.74 μg/kg and from 0.69 to 12.34 μg/kg, respectively. The RSDs were in the range of 1.8%-11.4% and 5.1%-13.9% for the single-needle and needle-to-needle methods, respectively. The proposed method was applied to soybean, peanut, and sunflower oils, and two PAEs were found in all three oils. Moreover, the method demonstrated good precision (RSD=1.17%-11.73%) and recoveries (72.49%-124.43%). Compared with other methods, the developed method was able to extract many target analytes and had a low or comparable LOD and high recovery. More importantly, this method does not require tedious operations such as solvent extraction and purification. Consequently, the developed method can be used to extract not only PAEs in oils but also other substances with a high lipid content.