code-review
🔌Plugingrixu/cc-toolkit
|Unified code review that fans out five parallel scanners over a change — comment quality plus four quality/craft lenses (readability & tests, naming & module, objects & patterns, simplicity & types) — and merges them into one per-file report. The bundled /start-cr command runs all five scanners; the standalone comment-review and quality-review skills stay invocable for a single-lens pass. Reviews the current branch diff by default (or explicit paths / --base). Quality findings are tagged family · rule · severity; comment verdicts are R1–R12 · KEEP/REMOVE/REWRITE/MOVE, shown side by side in one combined report and applied through a single risk-cut menu. Successor to the comment-review and quality-review plugins.
Part of
grixu/cc-toolkit
Installation
/plugin marketplace add grixu/cc-toolkit/plugin install code-review@cc-toolkitMore from this repository10
Claude Code plugin marketplace — productivity tools for AI-assisted development workflows, including plan review and requirements engineering
Intercepts ExitPlanMode to automatically review implementation plans via OpenAI Codex CLI. Includes /codex-review command for manual invocation.
End-to-end feature delivery: requirements gathering, editing, status tracking, and implementation orchestration with parallel subagents
Easily create hooks to prevent unwanted behaviors by analyzing conversation patterns
Analyze conversation for friction signals (corrections, repeated instructions, deviations) and interactively update project documentation (.claude/rules/, CLAUDE.md) with verified learnings. Coordinates with Claude Code's auto memory to avoid duplication.
Business name discovery: generates candidates, challenges them with a sub-agent, then checks domain availability via Instant Domain Search MCP.
Audio/video transcription pipeline with summary, news extraction, and custom processing. Sources: YouTube URLs (yt-dlp) or local files (any format ffmpeg handles). Backed by ElevenLabs Scribe.
[DEPRECATED → code-review] Single-skill code review focused exclusively on comment quality. Flags comments that narrate the code, banner/section dividers, change-state and ticket/PR history notes, cross-file references, process-narration disguised as decisions, commented-out code, and comments that contradict the code — and surfaces missing WHY at non-obvious code. Reviews the current branch diff by default (or explicit paths), with calibration against common false positives.
[DEPRECATED → code-review] Single-skill code review focused on code quality and craft, not correctness — twenty rules in seven families distilled from Kent Beck and Clean Code: readability (blank-line openness, guard clauses, explaining variables, named literals, composed methods, stepdown ordering), given/when/then test structure, naming (intent- and role-revealing names, command/query separation), module shape (OOP/functional consistency, pointless barrel exports), object & state design (fully-formed construction, lazy init, safe collection accessors), design patterns applied only under real friction (composition over inheritance, polymorphism over repeated conditionals, execute-around bracketing), and simplicity (over-complicated code that collapses into one parameterized function, unnecessary type casts). Reviews the current branch diff by default (or explicit paths), with calibration against common false positives.
Source-grounded research assistant: an interactive skill gathers a research brief, then launches a bundled Dynamic Workflow that fans out firecrawl retrieval (WebSearch fallback) into cited findings, gates rounds on coverage and contradictions, and renders a self-contained HTML report where every claim traces to a numbered source. The report evolves across follow-up runs.