plugins/cubic/skills/cubic-review/SKILL.md
Run AI-powered code reviews using Cubic CLI to detect bugs, security vulnerabilities, and style issues in local changes. Use when the user says "review my code," "check my changes for bugs," "run cubic review," "review this diff," "pre-commit check," "find issues before I push," "analyze my branch changes," or "code quality check." Triggers on mentions of cubic, code review, diff review, pre-commit checks, bug detection, and code quality validation.
npx skillsauth add pleaseai/claude-code-plugins cubic-reviewInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Detect bugs, security issues, and code quality problems in local changes by running cubic review via Bash.
Cubic CLI must be installed. If cubic is not found in PATH, inform the user with installation options:
curl -fsSL https://cubic.dev/install | bashnpm install -g @cubic-dev-ai/cliDo not run the installation command automatically. Let the user decide.
If cubic returns an authentication error, inform the user to run cubic login or complete browser-based authentication.
Always pass --json for structured output. Key modes:
cubic review --json -- review uncommitted changes (default)cubic review --base main --json -- compare against a branch (PR-style)cubic review --commit HEAD~1 --json -- review a specific commitcubic review --prompt "focus area" --json -- custom review focusConstraint:
--base,--commit, and--promptare mutually exclusive.
{
"issues": [
{
"priority": "P0",
"file": "src/api/auth.ts",
"line": 45,
"title": "SQL injection vulnerability in user lookup",
"description": "User input is concatenated directly into SQL query without parameterization."
}
]
}
Priority levels: P0 (critical) > P1 (high) > P2 (medium) > P3 (low).
cubic review --json (or with --base/--commit as appropriate)cubic review --json to verify fixesFor the full command reference and step-by-step workflow, use /cubic:review.
development
RTK (Rust Token Killer) reduces LLM token consumption by 60-90% by filtering and compressing command outputs. Use when user asks about token savings, token optimization, RTK usage, `rtk gain`, `rtk discover`, or `rtk proxy`. Triggers on mentions of token reduction, RTK commands, or checking command cost.
tools
Semantic code search using the Semble MCP server. Use when exploring an unfamiliar codebase, finding code by what it does rather than exact text, locating an implementation, understanding how a feature works, or discovering related code. Triggers on requests like "where is X handled", "find the code that does Y", "how does Z work", "search the codebase for", or any semantic/exploratory code question where grep's literal matching is a poor fit.
development
Get best practices for developing applications with Spring Boot.
development
Get best practices for JUnit 5 unit testing, including data-driven tests