Exploring RuboCop with MCP

This talk explores the potential of tools in the Ruby ecosystem in the AI era, with a particular focus on linters and formatters.

Traditionally, RuboCop has been triggered by humans or by other programs. In the AI era, AI agents have emerged as a new kind of trigger. This talk discusses practical ways to combine generative AI with linters and formatters, such as running these tools alongside AI agents, by integrating an MCP (Model Context Protocol) server implemented in Ruby.

While the AI landscape is evolving rapidly and technologies are changing at a fast pace, this talk aims to provide an opportunity for Ruby ecosystem developers to consider how existing Ruby tools like RuboCop can evolve in the AI era through MCP.

  • Koichi ITO
    Koichi ITO

    Koichi Ito is a member of the RuboCop core team, the MCP (Model Context Protocol) steering group, and an open source software maintainer. He is a long-time practitioner of Ruby and Rails application development with eXtreme Programming. He is also an Engineering Manager and Distinguished Engineer at ESM, Inc.