BACKGROUND: Pre-exposure prophylaxis (PrEP) is highly effective for HIV prevention, but uptake remains low, especially among adolescent girls and young women (AGYW) in Kenya. A model in which trained AGYW using PrEP deliver HIV self-tests to their close friends and refer them to PrEP may help increase PrEP uptake in this population. To understand AGYW's potential willingness to engage in such a model, we conducted a qualitative formative study in Kenya. METHOD: We conducted semi-structured in-depth interviews (IDIs) with AGYW (16 to 24 years) in Kiambu County who were determined at risk of HIV acquisition. We purposively recruited "PrEP-naïve" (no prior PrEP use) and "PrEP-experienced" AGYW who used PrEP for at least 1 month within the previous year. We solicited perspectives on initiating/engaging in conversations about HIV risk and PrEP, distributing/receiving HIV self-test kits, and referring/following through on a referral to clinic-based HIV services. We analyzed verbatim transcripts using rapid qualitative analysis and a combination of inductive and deductive approaches, with the latter informed by the Integrated Behavior Model (IBM). RESULTS: From August to December 2020, we conducted 30 IDIs: 15 with PrEP-experienced and 15 with PrEP-naïve AGYW. Participants' median age was 20 [interquartile range (IQR): 20-22]. Overall, most participants anticipated that they would be willing to engage in this model. PrEP-experienced AGYW emphasized the salience of their concerns about friends' HIV risk behaviors, with several noting that they are already in the habit of discussing PrEP with friends. Many additionally expressed positive attitudes toward the proposed target behaviors, perceived these to be normative among AGYW, and expressed confidence in their ability to carry out the behaviors with proper support. Although few participants had HIVST experience, nearly all anticipated they would be able to use an HIV self-test kit correctly if provided instruction. CONCLUSION: The Kenyan AGYW who participated in this study generally anticipated that they would be willing to engage in a formal peer PrEP referral model enhanced with peer-delivered HIV self-tests. Future research is needed to pilot test this model to determine its acceptability, feasibility, and effect on HIVST and PrEP uptake within this population.