Habitat management using flowering plants on rice bunds and field ridges surrounding rice paddies can enhance conservation biological control by supporting beneficial predatory insects. However, multi-year field evidence for sesame (Sesamum indicum) when deployed as these bund-side flowering strips, and there is no molecular gut-content confirmation that the key predatory mirid bug Cyrtorhinus lividipennis preys upon rice planthoppers pests under field conditions. Over a two-year field experiment, we compared rice paddies with sesame-planted bunds against control paddies without bunds. We surveyed major piercing-sucking pests (brown planthoppers Nilaparvata lugens, white-backed planthoppers Sogatella furcifera, leafhoppers, and thrips) and their primary predator C. lividipennis throughout different rice phenological growth stages. PCR-based laboratory digestion assays were used to quantify the detectability of N. lugens DNA in the gut of laboratory-reared and field-collected C. lividipennis. Sesame-planted bunds significantly increased natural-enemy abundance (C. lividipennis) and reduced planthoppers, leafhopper, and thrips densities. Treatment effects varied by rice stage, with suppression mainly occurring from booting/heading to grain filling and no significant differences at tillering, and effects at maturity were weak. In the laboratory, N. lugens DNA detection declined from 68.75% at 0 h to 0% at 16 h after feeding. In the field, 37.5% of C. lividipennis from sesame-bund paddies tested positive for N. lugens DNA. These findings provide field and molecular evidence that sesame-planted bunds conserve C. lividipennis and contribute to suppression of piercing-sucking rice pests, supporting their use as an auxiliary habitat-management tactic in rice integrated pest management.