This paper addresses the challenges in emotional generation tasks by proposing the Emotional Chain-of-Thought (ECoT), a plug-and-play prompting method that enhances the performance of Large Language Models (LLMs) in generating emotionally intelligent responses. ECoT aligns with human emotional intelligence guidelines, addressing issues such as human preference alignment and emotional generation assessment. To evaluate the effectiveness of ECoT, the authors introduce the Emotional Generation Score (EGS), an automated model-based evaluation method that incorporates Goleman's Emotional Intelligence Theory. Extensive experimental results demonstrate the effectiveness of ECoT and EGS, showing significant improvements in emotional generation tasks. The paper also discusses the potential applications of LLMs in emotional intelligence, such as emotional chat assistants and emotional rewriters, and provides key insights into the capabilities of LLMs with ECoT.This paper addresses the challenges in emotional generation tasks by proposing the Emotional Chain-of-Thought (ECoT), a plug-and-play prompting method that enhances the performance of Large Language Models (LLMs) in generating emotionally intelligent responses. ECoT aligns with human emotional intelligence guidelines, addressing issues such as human preference alignment and emotional generation assessment. To evaluate the effectiveness of ECoT, the authors introduce the Emotional Generation Score (EGS), an automated model-based evaluation method that incorporates Goleman's Emotional Intelligence Theory. Extensive experimental results demonstrate the effectiveness of ECoT and EGS, showing significant improvements in emotional generation tasks. The paper also discusses the potential applications of LLMs in emotional intelligence, such as emotional chat assistants and emotional rewriters, and provides key insights into the capabilities of LLMs with ECoT.