Aims: to develop and observationally evaluate a multi-component programme for the reduction of risk behaviours in vulnerable adolescents. Methods: the programme was theoretically informed by Catalano and Hawkins' social development model. It was developed using a combination of participatory consultation with young people and a review of evidence reviews. The resulting programme involved screening of school pupils at ages 14-16 years to identify those at risk, inviting them to attend two generic drug and alcohol awareness sessions, and then eight targeted life skills training sessions, alongside one-to-one motivational interviews and the creation of contacts with youth services. Evaluation was carried out using qualitative methods and the quantitative analysis of timeline follow-back questionnaires on drug and alcohol use at entry, exit and six-month follow-up from 226 programme participants.Findings: qualitative data showed that the programme was feasible, acceptable and positively viewed by participants, delivery staff and school staff. The professional skills of delivery staff were important in making it feasible and acceptable. Quantitative evaluation showed significant reductions in alcohol use (as measured by percentage days abstinent and drinks per drinking day). There were also reductions in illicit drug (mostly cannabis) use, although these were not statistically significant. Conclusion: the participatory development process produced a theoretically and evidence informed programme that was highly acceptable and appreciated by its target participants. This evaluation provides evidence of potential effectiveness that is worthy of further evaluation using more rigorous scientific approaches. The RisKit programme is available for use under a Creative Commons licence.