Claude Code plugins

Every plugin for Claude Code we can find on GitHub, read straight from each repo's .claude-plugin manifests. Install one with claude plugin marketplace add owner/repo and claude plugin install name@marketplace — each page has the exact lines. Looking for Codex? See Codex plugins.

karpathy-wiki

claudecodex

Auto-maintained LLM wiki based on Andrej Karpathy's pattern.

5 skills · 1 command · hooks

toolboxmd/karpathy-wiki 105

Search, read, organize, draft, templates, rules, and inbox stats for Apple Mail — with an IMAP fast path and a security-first design. macOS only.

5 skills · 1 MCP server

s-morgan-jeffries/apple-mail-fast-mcp 105

sudus

claudecodex

The referee for agent-led software development: reads the repository, names the next action.

5 skills · hooks

eas4ai/sudus 105

training-hub

claudecodex

LLM post-training — unified interface for SFT, OSFT, LoRA fine-tuning, and GRPO reinforcement learning

4 skills

Red-Hat-AI-Innovation-Team/training_hub 105

entrix

claude

Harness Engineering plugin exposing executable fitness functions for architecture quality, change-aware validation, and MCP-powered review context.

1 skill · 1 MCP server

phodal/entrix 105

Design comprehensive test cases using PICT (Pairwise Independent Combinatorial Testing). Analyzes inputs, generates PICT models with parameters, values, and constraints for valid scenarios using pairwise testing. Outputs the PICT model, markdown table of test cases, and expected results.

1 skill

omkamal/pypict-claude-skill 105

Audit and improve front-end usability using 15 established design principles. Acts as a Senior UX designer/engineer reviewing your interface end-to-end.

1 skill

mistyhx/frontend-design-audit 105

authsome

claude

Login once, credentials live in an encrypted local vault, a local proxy injects them into outbound requests so the agent's env never holds raw keys. 45 providers preconfigured (14 OAuth2, 31 API key).

1 skill

agentrhq/authsome 105

cellar

claude

Look up the public API of any JVM dependency (Scala 3, Scala 2, Java) from the terminal

1 skill

VirtusLab/cellar 105

A band above the prompt with where the run stands, a pane with the steps, their times, the documents and the tasks, and /speckit-tracker to switch specs. Works with stock Spec Kit. Read only.

alfredoperez/speckit-companion 105

Iterative decision simulation with a 5-persona council, adversarial red-teaming, and mechanical convergence. Applies Karpathy's autoresearch + LLM Council patterns to high-stakes business and strategy decisions.

1 skill · 11 commands · 5 agents

harshilmathur/autodecision 104

weaviate

claude

Search, query, and manage Weaviate vector databases. Includes database operations, cookbooks for building full-stack AI applications, and CLI commands.

2 skills · 9 commands

weaviate/agent-skills 104

Live web search, full-page content extraction, and cited research for Claude. Look up current docs, API references, changelogs, error fixes, and web data while you code, plus finance data and guided You.com API integration.

7 skills · 4 MCP servers

youdotcom-oss/agent-skills 104

Vue, Tauri, architecture, security, and MCP-debugging skills tailored to the oats desktop app.

6 skills · 1 MCP server

ariso-ai/oats 104

Make motion design videos in code: story, hooks and script in 10 languages, consent-gated TTS with word timings and captions, a style library, brand kits, HyperFrames scenes (Remotion optional), and self-checking QA with an independent critic.

7 skills

aykutucesai/motionforge 104

Skills for deploying and benchmarking vLLM with Claude Code

6 skills

vllm-project/vllm-skills 104

Professional skill and subagent creation with dual-mode workflow: 12-step fast mode and 15-step full mode with behavioral pressure testing and TDD integration.

1 skill · 3 commands

rfxlamia/skillkit 104

Read-only live monitor for Claude Code sessions and subagents (never starts or steers agents): logs Claude Code hook events locally and serves a live view on 127.0.0.1:4318.

2 skills · 1 MCP server · hooks

tomtomsatnav/agent-orchestration-standard 104

go-guidelines

claudecodex

Enhances code agents with Go best practices covering performance, modern syntax, generics, patterns, testing, error handling, and concurrency.

1 skill

mhmtszr/go-guidelines 104

imprint

claude

Your habits, imprinted on AI. One plugin replaces eleven skills: memory, compression, onboarding, code review, debugging, planning, progress tracking, testing, git workflow, SEO, and copywriting.

1 skill

ilang-ai/Imprint 104

Hypothesis property-based testing plugin

1 command

mmaaz-git/agentic-pbt 104

Record and transcribe audio to markdown using whisper.cpp with Metal acceleration. Captures mic + system audio.

1 skill

hrescak/transcribe-md 104

glycowork

claude

Use glycowork for any glycan-related coding task

1 skill

BojarLab/glycowork 104

Essential skills for everyday development: beginner-friendly explanations (baby-education), comprehensive README creation (readme-expert), LLM prompt optimization (prompt-engineering), AI-to-human doc transformation (humanize-docs), technical spec creation (quick-spec), pre-deployment QA checklisting (pre-deploy-checklist), and developer progress logging (been-there-done-that).

rfxlamia/skillkit 104

Agent capability frameworks: STAR framework for understanding implicit intent (framework-initiative), critical thinking architecture with reasoning selection (framework-critical-thinking), and adversarial review protocol with mandatory bug quota (adversarial-review). Build smarter, self-correcting agents and validate plans before execution.

rfxlamia/skillkit 104

summer

claudecodex

The game engine built for AI. Any agent makes a multiplayer game and publishes it. A first-run guide (/summer:start), game-dev skills, and two MCP servers: the bridge to the Summer Engine desktop app, and Summer's hosted cloud server for people without it.

138 skills · 3 commands · 2 MCP servers · hooks

SummerEngine/summer 103

ux

claude

v3.0, THE BRAIN. 18 callable slash commands (init, discover, design, system, audit, critique, a11y, copy, motion, polish, fix, lint, case-study, next, expert, research, workshop, mcp) plus 7 aliases kept until 4.1 (frame, recommend, stats, evolve, component, dashboard, image-to-code), 5 specialized sub-agents, 18 MCP tools, Python-backed 7-axis synthesizer with triple-mode dispatch (strict_brand / brand_anchor / pure_synthesis), 12 queryable JSON manifests (1,243 entries total), 152-rule deterministic anti-AI-slop linter with 0-100 quality scoring, mandatory 10-field discovery protocol, 160 brand DESIGN.md specs as training data (not templates), local decisions ledger that re-ranks the recommender from past wins, cross-IDE distribution to 17 environments (Claude Code, Cursor, Windsurf, Copilot, Gemini, Codex, Cline, Continue, Aider, Zed, JetBrains, etc.). Offline. Deterministic. No LLM. pip install uxskill or npx uxskill init.

1 skill · 25 commands · 5 agents · hooks

Laith0003/ux-skill 103

laoyeye

claude

戒指老爷爷 laoyeye。戒指里住着我的提示语老师们,31 件逐字收录。

32 skills

LearnPrompt/laoyeye 103

Skills adapted from Matt Pocock's public collection.

24 skills

mrclrchtr/supi 103

SuPi-owned public agent skills.

24 skills

mrclrchtr/supi 103