Plugin bundles without hooks
A hook is a shell command a plugin bundle runs on events of the AI tool, such as before a tool call or when a session starts, without you asking for it. The bundles in this list declare no hooks: they bring skills, commands, agents or MCP servers, and act only when you or the model call them. Bundles without a GitHub star figure come after the ranked ones.
Plugin bundles without hooks
Ranked by Stars (GitHub). Figures of different marketplaces are never added up.
-
Matt Pocock's agent skills for real engineering — grilling, spec/ticket flows, TDD, code review, domain modelling and more. Plug-and-play, not vibe coding.
GitHub stars: 285 737 -
Lazy senior dev mode. Forces the simplest, shortest solution that actually works: YAGNI, stdlib first, no unrequested abstractions.
GitHub stars: 161 091 -
Persistent memory system for Claude Code - seamlessly preserve context across sessions
GitHub stars: 99 711 -
Understand Anything is a Claude Code plugin that analyzes your project with a multi-agent pipeline, builds a knowledge graph of every file, function, class, and dependency, then…
GitHub stars: 85 924 -
Great design prompts require design vocabulary. Most people don't have it. You can't ask for "more vertical rhythm" if you've never used those words. Impeccable gives you commands…
GitHub stars: 79 844 -
Connect Mem0 to Claude to give your agent persistent memory across sessions. Claude can store user preferences, project context, and conversation history, then search, update, or…
GitHub stars: 66 990 -
AI-powered Draw.io diagram generation with real-time browser preview. Create flowcharts, architecture diagrams, and more through natural language.
GitHub stars: 36 172 -
Real-time statusline HUD for Claude Code - context health, tool activity, agent tracking, and todo progress
GitHub stars: 28 461 -
Teaches AI coding agents to create and maintain promptfoo eval suites. Encodes best practices like deterministic assertions first, file-based test organization, correct…
GitHub stars: 25 902 -
Skills for adding DeepEval evaluations, tracing, datasets, Confident AI reports, and iterative improvement loops to AI applications.
GitHub stars: 18 741 -
Open-source browser agent driven via CDP — direct browser control, 79K stars, YC W25. browser-harness gives Claude Code direct, low-level control of a real Chrome browser via the…
GitHub stars: 18 391 -
66 specialized skills for full-stack development: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6…
GitHub stars: 11 827 -
Build, train, evaluate, and use open source AI models, datasets, and spaces.
GitHub stars: 11 157 -
An agent skill that turns complex terminal output into styled HTML pages you actually want to read.
GitHub stars: 10 359 -
Query Anarlog meetings locally, or connect through OAuth to read opted-in Cloud snapshots.
GitHub stars: 9 475 -
Harness leverages Claude Code's agent team system to decompose complex tasks into coordinated teams of specialized agents. Say "build a harness for this project" and it…
GitHub stars: 9 155 -
Comprehensive development workflow for building AI agents with the Atomic Agents framework. Includes specialized agents for schema design, architecture planning, code review, and…
GitHub stars: 6 283 -
Skills for working with Daft, a high-performance data engine for AI and multimodal workloads. Includes UDF performance tuning, distributed scaling with Ray, and documentation…
GitHub stars: 5 795 -
Exa AI web search, deep research, and content extraction. Provides MCP tools and research skills for comprehensive web search, people discovery, company research, academic papers…
GitHub stars: 5 106 -
Database management tools for Claude Code. Query databases, explore schemas, analyze data quality, and get optimization recommendations. Supports PostgreSQL, MySQL/MariaDB…
GitHub stars: 5 037
How the list is built
- Type: Plugin bundles.
- Declares no hooks: nothing runs on events of the AI tool by itself.
- Ranked by Stars (GitHub); plugins without this figure come after the others and are marked, never shown as zero.
- GitHub figures rank a plugin only when its repository holds no other plugin we track; the figure of a shared repository is shown as the whole repository's and does not rank.
- Archived plugins are left out: an archived repository, or a marketplace that marks the plugin archived.
- Only known plugins whose page we index: listed on a curated marketplace or on two marketplaces or more, or with at least 50 stars on GitHub (of its own repository), 500 installs on Claude Marketplace or 5 000 pulls on Docker Hub.
- The list shows up to 20 plugins and is offered to search engines only with 8 or more.
The list is rebuilt from our latest passes. Nobody pays to appear in it or to rank higher.
All AI tool plugins · AI tool plugin statistics
Where the data comes from: how the catalog is built