Skip to content
CAI
Software that uses CAICheck a score

jtgsystems/Custom-Modes-Roo-Code

64.1

Adequate · 9 November 2026

3.6k

lines of production code

Python

primary language

10

measurements over time

CAI band scale
CAI trend line
CAI lens gauges

What this system is

This system is a configuration management tool for Roo Code that standardizes and automates the creation of custom AI agent personas. It provides a structured framework for defining specialized roles, such as Accessibility Experts or CLI Developers, through YAML-based mode definitions. The system includes automated scripts to validate, merge, and inject specialized content into these configurations, along with utilities to convert and deploy these modes for both CLI and VS Code environments.

Features

Add CLI tool to convert and manage Roo Code custom modes for VS Code

Introduces a Python script (\convert\_modes.py\) and documentation that allow users to convert CLI-format custom modes into VS Code-compatible YAML and XML rule files. The tool supports listing, searching, and converting modes with smart merge capabilities (updating existing, adding new, preserving others), and includes commands to copy the generated configurations to local or remote VS Code settings directories.

vs-code · high confidence

Automated custom mode management and content injection tooling

Added a suite of Python scripts to automate the lifecycle of Roo Code custom modes. \compile\_modes.py\ reads individual YAML definitions from \custom\_modes.d/\, validates them against a strict schema (checking slugs, required fields, and permissions), and merges them into consolidated \.roomodes\ and \custom\_modes.yaml\ files. \generate\_roomodes.py\ provides flexible generation with support for batching, slug filtering, and size limits. \fix\_modes.py\ automatically improves mode quality by expanding short role definitions and resolving placeholder instructions. Additionally, \inject\_debugging\_playbook.py\, \inject\_engineering\_protocols.py\, \inject\_gap\_skills.py\, and \inject\_github\_superpowers.py\ programmatically inject specialized content—such as SOTA 2026 debugging protocols, engineering rules, and external skill sets—into specific persona configurations.

scripts · high confidence

Introduction of custom AI agent modes and standardized repository configuration

The repository now supports a system of custom AI agent modes, defined in the new \.roomodes\ file (and its variants), which provides specialized personas such as Accessibility Expert, AI Engineer, and CLI Developer with detailed instructions and tooling guidelines. To support this structure, \.gitattributes\ has been added to enforce LF line endings across all text-based configuration and code files, and a comprehensive \.gitignore\ has been introduced to exclude Python build artifacts, IDE settings, and local environment files.

(repo-wide) · high confidence

Written by cca.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

How this codebase got here

Score

  • CAI 58 → 64 (+5.9)
  • Rubric changed (rubric-2026.10.5 → rubric-2026.11.2) — scores are not directly comparable.

Lenses

  • Code Health 63 → 68 (+5.4)
  • Architecture 100 → 100 (+0.0)
  • Maturity 51 → 68 (+17.3)
  • Readiness 54 → 54 (+0.0)
  • Security 85 → 85 (+0.0)

Resolved (14)

  • Documentation: contradicts the code (docs/researched.md)
  • compile_modes.load_existing_roomodes (cognitive 18) (scripts/compile_modes.py)
  • compile_modes.load_individual_modes (cognitive 20) (scripts/compile_modes.py)
  • compile_modes.main (cognitive 19) (scripts/compile_modes.py)
  • compile_modes.validate_mode (cognitive 25) (scripts/compile_modes.py)
  • compile_modes.validate_mode (cyclomatic 21) (scripts/compile_modes.py)
  • fix_modes.main (cognitive 19) (scripts/fix_modes.py)
  • generate_roomodes.main (cognitive 24) (scripts/generate_roomodes.py)
  • inject_engineering_protocols.inject_into_file (cognitive 27) (scripts/inject_engineering_protocols.py)
  • inject_github_superpowers.main (cognitive 17) (scripts/inject_github_superpowers.py)
  • migrate_all_agents.main (cognitive 23) (scripts/migrate_all_agents.py)
  • trim_modes.main (cognitive 18) (scripts/trim_modes.py)
  • verify_modes.verify_mode (cognitive 23) (scripts/verify_modes.py)
  • verify_modes.verify_mode (cyclomatic 20) (scripts/verify_modes.py)

Written by cca.canine.dev from the codebase's own history, inside the signed delivery this page is composed from.

Survey your own repository

jtgsystems/Custom-Modes-Roo-Code was measured the same way every project in this corpus was: the same rubric, at a pinned commit, with the result published in full. Point a surveyor at a repository you know and see whether you agree with it.

About this page

  • The score is its most recent published measurement, taken on 9 November 2026 at a pinned commit. It is not a live figure and does not change until the project is measured again.
  • Measured at commit bd349c4016011717251b5ea53c655eee7b2d649e — the exact code this score is about.
  • Scored under rubric-2026.11.2 — the same rubric and the same method as every other entry in this index.
  • Measured by Canine Code Analyzer (cca.canine.dev) using cca measure-2026.11.4+preprod-464a0040a931.