The objective of this research is to evaluate the feasibility of an automated intervention delivered via a social media-based chatbot to reduce online victimization detected in primary care and prevent depression and suicidal thoughts and behaviors (STBs) among adolescents with subsyndromal depressive symptoms, with special attention to the needs of Black and sexual and gender minority (SGM) youth. Online victimization (OV) (i.e., disparaging remarks, symbols, images, or behaviors that inflict harm in online spaces) frequently occurs on social media, and is directly linked to depression and suicidal ideation, which in turn associate with near-term suicide risk. OV peaks during adolescence and it is especially prevalent among minority adolescents; over half of Black youth experience a form of race-based online victimization involving derogation or exclusion, and SGM youth experience OV at three times the rate of non-SGM youth. Early detection and intervention on OV has potential to improve health during adolescence and prevent longer-term negative effects. Primary care providers (PCPs) recognize that OV is common among their patients, but practical barriers inhibit their capacity for response. Chatbots are software programs that use artificial intelligence to simulate conversation with a user via text, voice, or video, and can be deployed on social media, SMS, or can be web-based. A chatbot that responds to OV identified in primary care could extend PCPs’ capacity to help youth with OV through an automated intervention. We propose to develop and evaluate SMILEY, a chatbot-delivered intervention deployed on social media that will reduce OV frequency and intensity of distress (factors associated with depressive severity and suicidal ideation) among online-victimized adolescents with subsyndromal depressive symptoms identified in primary care. To inform intervention development, we will apply the Consolidated Framework for Implementation Research (CFIR) to the conduct of qualitative interviews to identify perceptions and preferences regarding a chatbot-delivered intervention for online victimization. We will then use human-centered design methods to develop, evaluate, and iterate SMILEY. We will then conduct a randomized trial comparing SMILEY vs. a brief psychoeducational intervention among recently victimized adolescents with subsyndromal depression. Together, these aims will provide data regarding a novel chatbot intervention that may hold promise to fill a critical gap in services for adolescents experiencing online victimization and meaningfully advance suicide prevention for vulnerable youth. We aim to develop an intervention that is acceptable and feasible for minoritized youth, specifically Black and SGM youth, who are especially vulnerable to experience online victimization, thus addressing a source of health disparities in these youth.