FRC Code Scout

Scores FRC/FTC team code against an 8-dimension sophistication rubric and helps teams build elite robot architecture.

FRC Code Scout is an AI-agent-driven tool that does two things for FRC (and FTC) robot code:

  1. Scores a team’s code. It runs your (or any team’s) robot repo against an 8-dimension code-sophistication rubric — architecture, coordination, simulation, testing, logging, autonomous, vision, and sustainability — and benchmarks the result against a 24-team San Diego survey. You get a candidate score per dimension, the evidence behind it, and a prioritized list of what to improve next.
  2. Builds robot code the elite way. It scaffolds new robot projects with the same architecture the rubric measures: a hardware IO layer, sim-backed testing, structured logging, and the other practices that separate elite teams from the rest. You can start a new project on this foundation, or add one subsystem at a time to an existing codebase.

Why you’d come here

  • See how your team’s code compares to dozens of other San Diego-area FRC teams, dimension by dimension.
  • Get a prioritized improvement list instead of a generic linter report — the tool confirms every finding by reading the actual code, not just checking for the presence of a library.
  • Build your robot code the elite way — scaffold the IO layer, simulation, logging, and testing seams used by top teams, or add them incrementally to code you already have.

Learn more

The full project — the rubric, the research corpus, the build-spec book, and the tool itself — is published at:

https://robots.jointheleague.org/frc-code-scout/