Code-Quality plugins

A constructive critic that gives a second opinion on your own plans, designs, code, and tests — spawning independent read-only reviewers on different Claude models than the one driving the session

1 skill

Teach your AI coding agent the MASA (Modular Agentic Semantic Architecture) framework

2 skills · 6 commands

Review Before Merge workflow: change-review and security-review skills, a security-reviewer subagent, a post-edit test hook, and an unsafe-command blocking hook.

2 skills · 1 agent · hooks

Hook PreToolUse que bloquea los comentarios nuevos en ficheros Ruby (.rb/.rake): el código se explica solo. Exime los pragmas de máquina (frozen_string_literal, sorbet, rubocop).

hooks

Hook PreToolUse que bloquea los comentarios nuevos que el agente añade al código, en cualquier lenguaje: el código se explica solo. Exime los pragmas de máquina (noqa, eslint-disable, @ts-, shellcheck, licencias).

hooks

20 specialized agents and 9 commands for exploring, analyzing, and improving existing Python codebases. Includes analysis scripts for correctness and concurrency scanning against a bug-shape catalog, tiered lint triage, mypy integrity checks, import graphing, complexity measurement, test correlation, type coverage, dead symbol detection, tech debt collection, git history analysis, web-framework wiring and ORM-schema checks, a Phase 3 publication pipeline, and threat-model discovery with a security-claims gate.

9 commands · 20 agents

Rules your coding agent physically cannot ignore. A Claude Code plugin (and an LLM-agnostic rule pack) that intercepts the agent's own tool calls instead of asking it nicely: PreToolUse denies an edit that skips read-before-edit, plants an unjustified suppression marker, hardcodes a secret or a machine-specific path, or piles rolling patches onto one file; a Bash gate refuses --no-verify, force-push and similar bypasses; and a nine-layer Stop gate refuses to let a completion claim end the turn without evidence, a convergence self-quiz, a fidelity check against the original request and a one-line TL;DR. Twelve rules (ten hook-backed), per-project Imperial Edict regexes you define yourself, a repo-wide sync gate, and CI drift gates that derive every documented number from the code. Standard library only; every guard fails open, so a bug in the discipline can never brick your agent. Release history: CHANGELOG.md.

2 skills · 6 commands · 1 agent · hooks

Deep analysis of codebases to identify duplicate code, unused exports, dead code, and unnecessary dependencies

1 command · 1 agent

Simplifies and refines code for clarity, consistency, and maintainability while preserving all functionality

1 skill

Pre-commit expert that discovers, runs, and resolves all code quality checks in a monorepo

1 skill

Analyze Mattermost schema migrations against best practices and generate a review report

1 skill

Scheduled maintenance for AI codebases. Catches the drift reviews miss and lines it up.

1 skill · 1 command · 3 agents

An independent quality gate for AI-generated code. A different model reviews what the agent just wrote — against your project's standard — before it lands.

Deep code review with 23+ specialist Opus agents. Full call-graph tracing, web-verified findings, adversarial false-positive filtering. Supports GitHub PRs, GitLab MRs, and local branch comparisons.

1 skill · 25 agents

Executive team perspectives for any codebase — CEO, CTO, PM, CDO, CSO, CLO, COO, CMO, CAIO, CFO, CIO, QA Lead, DX Lead reviews

15 commands

Code review skills and agents for catching bugs, security issues, and style problems

1 skill · 1 command · 1 agent

Ask the compiler index before deleting a Swift declaration: who uses it, what it uses, whether it is reachable, and what the analysis cannot see.

1 skill

13 specialized agents and 4 commands for analyzing PyO3 Rust extensions. Tree-sitter-rust scripts detect unsafe-block soundness violations, panics crossing the FFI boundary, swallowed PyResults, #[pyclass] Send/Sync gaps, missing __traverse__/__clear__, and PyO3 version-migration debt. Includes a 'migrate' command for the Bound-API and free-threading migration checklist.

4 commands · 13 agents

AI-powered code review and PR comment resolution with CodeAnt AI

10 skills

Claude Code marketplace wrapper for AgentLint's AI-agent guardrail engine. 77 rules across 8 packs, hybrid AgentChute security feeds, circuit-breaker visibility, inline-ignore reasons.

2 commands · 3 agents · 1 MCP server · hooks

Semantic refactoring opportunity detection. Scans committed code for structural duplication (Fowler catalog), code smells (Fowler/Beck), GoF design pattern opportunities, SOLID/DRY principle violations, and language-idiomatic anti-patterns (Go, Python, TypeScript). Creates beads issues with TDD-first refactoring plans. Run on demand with /refactor:scan.

1 skill · 3 agents

Tornhill-style hotspot and temporal-coupling analysis on any git repo, turned into one design improvement: a long-term goal plus the smallest meaningful step toward it.

3 skills

Distils a project's review history, docs and enforced configs into a repo-local review skill that carries only the delta — where its maintainers do not want what a default reviewer would ask for.

3 skills

Ready Agent 1 — is your codebase ready for the agents? Score a repository's agent-readiness (deterministic, cited, Level 1–4 with Level 5 Autonomous reserved) and clear the defined gates to raise it.

3 skills

Improves Python code quality using PEP 20 (Zen of Python): clearer structure, readability, and maintainability. Guides writing and reviewing any Python — programs, scripts, functions, one-liners — plus snippets, bug fixes, refactors, and code reviews. Triggers on Python work, PEP 20, style, or “is this good Python?”.

1 skill

Reviews code changes for quality, correctness, style, and potential issues.

1 skill

Audits dependency files and package changes for vulnerable packages, risky upgrades, license concerns, and supply-chain exposure.

1 skill

Plans safe framework, library, API, database, or architecture migrations with phased rollout, compatibility checks, and rollback strategy.

1 skill

12 specialized agents and 16 commands for reviewing SHAP.

24 commands · 12 agents

Executive team perspectives for any codebase — CEO, CTO, PM, CDO, CSO, CLO, COO, CMO, CAIO, CFO, CIO, QA Lead, DX Lead reviews

16 skills · 16 commands