17,691 mods in this category, of every kind an agent can take. Each one carries what it costs per session, what the scan found, and whether it is the original.
Kenmege/codex-claude-companion
Skill Claude Code
Use when the user asks Codex to orchestrate Claude for a coding job in another terminal, or to review code, run adversarial or ship/no-ship review, audit security risks, or inspect a large diff/folder.
cmersman/security-review-plugin
Plugin Claude Code
Bundles 1 agent · 66 tokens together
Security-focused code review agent for identifying vulnerabilities, auditing authentication logic, and ensuring OWASP compliance.
penkzhou/swiss-army-knife-plugin
Plugin Claude Code
Bundles 11 skills, 9 commands, 2 hooks · 683 tokens together
Multi-stack workflow plugin with unified 3-layer architecture. Features 6-phase Bugfix, 8-phase PR Review, 7-phase CI Job Fix, and 6-phase Execute Plan workflows. 47 agents including 4 master-coordinators and shared review-coordinator for fully closed-loop execution.
ToshikiImagawa/unity-claude-code-tools
Agent Claude Code
A design reviewer checks Claude Code plugin sub-agents against guidelines written in AGENTS.md. A sub-agent is a helper agent that performs a focused task inside a larger plugin.
Plugin Claude Code
Bundles 3 skills, 7 commands, 1 agent, 3 hooks · 99 tokens together
Use Gemini CLI from Claude Code to review code or delegate tasks.
Plugin Claude Code
Bundles 1 skill, 1 command, 2 agents · 1,262 tokens together
Autonomous development loop: plan, ship, run codex adversarial-review, and iterate on findings until approved or threshold/iteration cap is reached. v0.3.2 adds --from-pr [ ] — a first-class entry path for mid-development PR handoff. When plan/implement/ship were done outside clodex, --from-pr bootstraps .clod.
Plugin Claude Code
Bundles 4 skills · 86 tokens together
A generator kit for Claude Code dev tooling. Three generators read your codebase and write a project-specific review skill, a rules set, and a self-learning feedback loop tuned to your stack, instead of handing you tooling written for someone else's project. The review skill they write is an overlay on a shared engine.
Xandon/claude-filemaker-toolkit
Plugin Claude Code
Bundles 1 skill, 4 commands · 250 tokens together
Analyze FileMaker Pro DDR XML exports — index scripts, fields, layouts, relationships, and custom functions into SQLite for fast querying, code review, dependency analysis, and paste-ready implementation packages for scripts, layouts, and schema changes (FM 18+ clipboard format).
Command Claude Code
Run a context-aware pull request review using multiple specialized agents, each focusing on a different aspect of code quality. This is a read-only review — do NOT make code changes or post GitHub comments unless explicitly asked.
Plugin Claude Code
Bundles 5 skills, 1 command, 7 agents, 3 hooks · 594 tokens together
A general spec-driven software development workflow for Claude Code: detect project domain/conventions, research, weigh solution options, plan, implement in reviewed phases (code review every phase; a security scan whenever a phase touches a sensitive surface), and run a final cross-phase security audit. Stops for…
Command Claude Code
Open an interactive code review for current changes.
Plugin Claude Code
Bundles 7 skills, 2 agents · 353 tokens together
AI-powered code review and visualization — analyze, visualize, and improve any codebase.
Plugin Claude Code
Bundles 7 skills, 6 commands, 3 agents, 4 hooks · 774 tokens together
Jintech OMG development toolkit — SDLC skills (/ticket, /implement, /verify, /prepr, /pr, /grill-me, /debug) + code-review-graph & product-graph MCP servers + autonomous verify trust core + enforcement hooks.
Cursor rule Cursor
Cursor rule "clean-code" from ChiragAgg5k/poke-mcp, covering clean code guidelines, constants over magic numbers, meaningful names, smart comments and single responsibility.
arjunthilak05/antigravity-claude-mcp
MCP server Claude CodeCodexCursor +2
Use Google Antigravity (Gemini 3 Pro, Claude, GPT-OSS) as an independent second-opinion / code-review model inside Claude Code — via MCP, one command, zero dependencies. Runs locally from the antigravity-claude-mcp npm package. Needs 2 environment variables to run.
Skill Claude CodeCodex
Structured code review workflow and security checks.
yingsf/pycharm-code-quality-mcp
MCP server Claude CodeCodexCursor +2
MCP server "pycharm-code-quality-mcp" as configured in yingsf/pycharm-code-quality-mcp. Runs locally from the pycharm-code-quality-mcp Python package.
Cursor rule Cursor
CodeRabbit provides code review capabilities in this workspace.
Plugin Claude Code
Bundles 2 skills, 1 command, 1 MCP server · 147 tokens together
Collaborative code review CLI — browse diffs, leave structured comments, and save review reports. Token-efficient alternative to the MCP server: agents invoke batch-review via their shell tool instead of loading tool schemas into context.
shaunie2fly/ndlovu-code-reviewer
MCP server Claude CodeCodexCursor +2
One of 2 in .mcp.json
MCP server "ndlovu-code-reviewer" as configured in shaunie2fly/ndlovu-code-reviewer. Runs /mnt/nvme1-data/home/shaunbotsis/development/ndlovu-code-reviewer/dist/index.js with node.
MCP server Claude CodeCodexCursor +2
MCP server "ndlovu-code-reviewer" as configured in shaunie2fly/mjmlmcp. Runs /mnt/nvme1-data/home/shaunbotsis/development/ndlovu-code-reviewer/dist/index.js with node.
AleBrito124356/mcp-devils-advocate
MCP server Claude CodeCodexCursor +2
MCP server "mcp-devils-advocate" as configured in AleBrito124356/mcp-devils-advocate. Runs locally from the mcp-devils-advocate Python package. Needs 1 environment variable to run.
MCP server Claude CodeCodexCursor +2
MCP server "cejel" as configured in BargLabs/cejel. Runs in Docker (ghcr.io/barglabs/cejel@sha256:f1cd2356cc0072ad08e24fcddc379206ee07f5699b9a633164326917f665b6d6).
Plugin Claude Code
Bundles 9 skills, 2 hooks · 461 tokens together
Project governance and engineering coaching for AI coding agents: align requirements and interface design, reuse frameworks and components, improve code quality, and deliver with evidence.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: