Background: The use of herbal medicines is justified empirically using ethnopharmacological knowledge, which has limitations. Reports of adverse events (AE) may contribute for safety, quality, and effectiveness assessment. Objectives: This study aimed to develop an instrument to allow detection and evaluation of the causal likelihood of AE related to herbal medicines. Methods: A quantitative methodological development research intervention was performed with primary care health professionals. Two cases were distributed in an interval of one week in order to allow for completion of the proposal form. The within-rater reliability was calculated and classified into satisfactory, regular, and unsatisfactory, according to essential, necessary, and recommended items reported. Results: Fifty-seven professionals were enrolled. A high degree of the within-rater reliability was observed for the most essential, necessary, and recommended fields. The instrument contributed to assessment of definite causal association, once the duration of reactions including evolution, withdrawal, and reposition items showed high reliability. Most fields poor and weak filled were related to quality defects, such as popular name and appearance. Conclusions: Data suggest the instrument is adequate to report AE arising from herbal medicines. Owing to information regarding to organoleptic characteristics, the instrument is different from drug instruments. Therefore, it is intended to improve AE reports related to herbal medicines and contribute for appropriate use.