Code-Quality plugins

The review gate as a tool.check function hook: every shell call decided in process, the script spawned only for a command that could commit, and a blocked commit put to you as a question rather than refused. Needs CLAUDE_CODE_ENABLE_FUNCTION_HOOKS=1 and bin/review-gate installed.

Score agent-facing tool descriptions 0-5 on discoverability. Takes a tool description, explains failures against 5 criteria, and rewrites it to pass.

Tool · Executive-team review lenses (CEO/CTO/CSO/CFO/PM/QA-Lead/DX-Lead/…) for any codebase. Single-lens /ask router plus per-role commands. (JFK-maintained; kagura-ai migration deferred.)

16 skills · 16 commands

AI code review for Claude Code and Codex. Reviews every PR, fixes issues via MCP, you decide what merges.

8 skills · 6 commands · 1 agent · 1 MCP server · hooks

Code that runs true under agentic maintenance. Ships the methodology rules (manifesto, style guide, cheatsheet), an idempotent affirm skill, a multi-language lint binary (comment hygiene with structural exemptions only, citation resolution), a PostToolUse hook, and a suite of compliance subcommands and skills.

13 skills · hooks

Scored code quality assessment across 10 dimensions with weighted composite scoring and quality gates

10 skills · 1 command

Analyze codebases to discover bugs, security vulnerabilities, architectural problems, and improvements, then create GitHub issues

2 commands · 7 agents

Tool · PhD-level review panel (CS / DB / Distributed-Systems / Data-Science / PL / Security / Statistics) for deep technical critique. (JFK-maintained; kagura-ai migration deferred.)

9 commands

Bundles the Qualimetry skills, a PostToolUse hook that surfaces review findings the moment a reviewed file is opened, and the qualimetry-setup workflow for registering the Qualimetry MCP server.

6 skills · 1 command · hooks

Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification

1 command · 6 agents

NASA Power of 10 rules as Claude Code skills that apply while writing and reviewing Go, Python, and C: adapted profile for application code, strict literal profile on request, plus a review command and agent.

4 skills · 1 command · 1 agent

Catches function name drift, phantom imports, field mismatches, and contract violations in AI-generated code.

3 skills · 1 agent · 1 MCP server · hooks

Zero-tolerance multi-agent code review

1 skill · 5 agents

Lazy senior dev mode whose reuse rung searches the whole VPS before writing new code

6 skills

플러그인 마켓플레이스 프로젝트의 구조 일관성, 문서 품질, 설정 정합성을 자동 분석하고 GitHub Issue로 등록하는 플러그인

2 skills · 4 agents

Audits this marketplace against its own F1–F13 failure classes: deterministic pure-shell/awk regression detectors (fail-open, path traversal, broken hook contracts, manifest + vendored-lib drift, notebook/stdin/timeout/line-ending gaps) plus breadth+depth review sub-agents. Writes .claude/auditor/FINDINGS.md and records a HIGH count in the STATUS contract so the pipeline release gate enforces 0-high. Silent on a clean tree; no python dependency.

2 skills · 1 command · 2 agents · hooks

Reviews a CHANGE (the working-tree diff vs HEAD) against the explorer MEMORY.md invariants and risk map, the approved PLAN scope, and the codebase's surviving callers. Deterministic pure-shell/awk checks — R1 caller-integrity & R2 convention-regression (BLOCKING), R3 risk-touch & R4 scope-discipline (CONCERN) — plus breadth+depth review sub-agents. Writes .claude/reviewer/REVIEW.md and records a BLOCKING count in the STATUS contract so the pipeline release gate enforces 0-blocking. Orthogonal to the auditor (change-vs-invariants, not whole-tree regression). Silent on a clean/in-spec change; no python dependency.

2 skills · 1 command · 2 agents · hooks

Records refactoring opportunities as Claude works, surfaces them at the end of each session, and addresses them with focused single-item agents.

1 skill · 2 commands · 1 agent · hooks

Inline code review with MCP integration — add comments in the browser via SSH port forward, let Claude Code read and fix them via MCP tools.

3 skills · 1 MCP server · hooks

Blocks new prose comments that don't pair into rhyming anapestic couplets (AABB).

3 skills · hooks

Lightweight adversarial code review with three parallel subagents (Advocate, Skeptic, Architect) that analyze independently before synthesis

1 skill · 3 agents

Puts your repo's quality norms into the agent's inner loop: a session-start brief, mid-edit nudges as edits reach a norm, and one bounded soft nudge at stop. Guidance, never a merge gate.

1 skill · 1 command · hooks

Blocks new prose comments that don't scan as iambic pentameter — every comment line must be exactly ten syllables.

2 skills · hooks

Blocks new prose comments that don't fit 5-7-5 haiku form inside /** ... */ blocks.

2 skills · hooks

Two read-only reviewers that run before an agent may declare a task done: verification-critic (reward-hacking and spec drift) and architecture-reviewer (structural decay). Optional Stop hook refuses to end a turn on red Tier-1 gates.

2 agents · hooks

Language-agnostic methodology for writing and refactoring code against SOLID, GoF design patterns, and complementary principles (DRY, KISS, YAGNI, Boy Scout, SoC, Encapsulation, Law of Demeter, CQS, TDD, AHA) — for whole-worktree refactors and live in-flight review alike. Wires into whatever CI/QA a repo already has and verifies every finding through an adversarial reviewer/verifier subagent pair before touching code.

1 skill · 2 agents

Always-on "write the least code that fully works" discipline. A SessionStart hook injects the full decision LADDER (skills/minimal-code) and a UserPromptSubmit hook re-injects a compact reminder every turn, so the agent ships the SMALLEST code that satisfies the requirement WITHOUT cutting validation, error handling, security, or accessibility. Both hooks are node-guarded and fail-quiet (node is OPTIONAL — absent means the hooks stay silent and the skill still works on demand) and INJECT CONTEXT ONLY (no gate decision; every gate in this marketplace stays pure shell/awk). Intensity off|lite|full|ultra via /minimize, mirrored into STATUS.json. YAGNI applies only to UNREQUESTED scope — anything the builder PLAN or the user mandates stays in-scope. Ladder adapted from Ponytail by Dietrich Gebert (MIT).

1 skill · 1 command · hooks

Detect test doubles and unapproved fallbacks in production code. AI coding tools silently introduce mock/stub/fake objects and fallback behaviors — this plugin catches them.

1 skill · 1 command · hooks

Teach appropriately-expressive backend code (Python/TS/Go): when to comment, abstract, validate, and handle errors. Every rule has a when-NOT guard, so it curbs LLM over-commenting/over-abstraction without overcorrecting into cryptic code. Bundles real before/after examples and a tripwire scanner.

2 skills

Review the diff of the current git branch against a target branch. Ships two flows: the multi-agent review-code-change skill (extensible standards — security, performance, business logic, React/Next.js/Node/TypeScript) and the /review-code slash command (blast-radius mapping grounded in real test/typecheck/lint output). Read-only git.

1 skill · 1 command