The 5 Best Claude Code Marketplace Alternatives in 2026

Anthropic's official plugin directory for Claude Code. Plugins are packaged bundles that can include skills, agents, hooks, MCP servers, and other capabilities, replacing the old scatter of CLAUDE.md files, hook scripts, and custom slash commands. The marketplace is automatically available when you start Claude Code, so there's nothing to install before browsing. You open it with the /plugin command, use the Discover tab, and install directly, for example /plugin install github@claude-plugins-official. Anthropic manages the repository (claude-plugins-official, around 32k stars) and also publishes a separate Knowledge Work Plugins collection aimed at knowledge workers in Claude Cowork. The catalog leans developer-first: LSP plugins for TypeScript, Python, Rust, and Go; MCP integration plugins for GitHub, Linear, Slack, Vercel, Sentry, and Supabase; code review and security plugins. The catch is that it only serves Claude Code, and using it well means paying for Claude Code, where the Pro plan gets tight once you lean on agents heavily.

pricing 20/month

• + Zero setup: the marketplace is automatically available when you start Claude Code
• + Curated by Anthropic, so the core catalog is high quality and well documented
• + Plugin packaging replaces scattered CLAUDE.md files, hook scripts, and MCP configs
• + Strong developer catalog: LSP, MCP integrations, code review, and security plugins

• − Requires a paid Claude Code subscription; Pro at $20/month gets tight with heavy agent use
• − Claude Code only: no support for Codex, Cursor, Gemini CLI, or chat-app agents
• − No published security scanning, license checks, or per-skill quality benchmarks
• − Installing many MCP plugins can choke the LLM context window as all tools load into context

https://github.com/anthropics/claude-plugins-official

SkillzBench is AIQ Labs' free, independent directory of AI agent skills plus a free MCP tool that connects any agent to the skills you choose. Agent skills (SKILL.md instruction packages for Claude Code, Codex, Cursor, Gemini CLI, and chat-app agents) have exploded across GitHub, but quality and safety vary wildly: published research found roughly 26% of skills carry at least one vulnerability. SkillzBench tackles that with five pillars. Discover continuously crawls and dedupes skills from GitHub. Verify pins every version to a commit SHA, validates it, checks the license, runs a security scan, and judges quality. Benchmark tests every skill in a sandbox with the same model, with and without the skill, and publishes the Lift in percentage points. Equip gives you one free MCP server plus per-agent loadouts, so you toggle skills and tools per agent and the agent only sees what you chose. And it stays current, tracking upstream changes, re-scanning, re-benchmarking, and notifying you. A 0-100 trust score combines security, quality, license, maintenance, and community signals; Lift is always shown separately, and money never touches scores. Everything on the platform is free.

pricing 0

• + Completely free, including the MCP connector and benchmarks; ad revenue never influences scores or listings
• + Security scanning and version pinning address the ~26% of skills research found to carry vulnerabilities
• + Works across Claude Code, Codex, Cursor, Gemini CLI, and chat-app agents, not just one tool
• + Sandbox Lift benchmarks show whether a skill actually helps, measured the same way for every skill
• + Per-agent loadouts keep agents lean, avoiding the context-window bloat of loading every tool at once

• − Focused on skills and their verification, not on packaged plugin bundles with hooks and commands
• − Smaller brand recognition than Anthropic's official marketplace
• − Trust scores and Lift results require reading them, some effort versus one-click official installs

https://skillzbench.com

Maintained by Composio, this is a curated GitHub collection of Claude Code plugins organized by what they do, and one of the best-known third-party marketplaces. You add it to Claude Code with /plugin marketplace add owner/repo-name, or clone the repo and install directly. Its strength is editorial judgment: rather than dumping every plugin into one list, Composio's guide sorts picks by type, code intelligence and LSP plugins (TypeScript, Pyright, Rust Analyzer, Go), MCP integration plugins (GitHub, Linear, Slack, Vercel, Sentry, Supabase), code review plugins (Code Review, PR Review Toolkit, CodeRabbit, Code Simplifier), and security plugins (Security Guidance, Semgrep). The accompanying write-up gives practical sequencing advice, like installing one LSP plugin for your main language before adding anything more complex. It also calls out real caveats, such as multiple MCP servers choking the LLM context window. It's Claude Code focused and doesn't publish per-skill security scans or benchmarks, so curation is human judgment rather than measured testing.

pricing See pricing

• + Human-curated with clear recommendations per category rather than a raw dump
• + Backed by a detailed, practical guide on what to install first and why
• + Composio Connect MCP consolidates app access into one server, easing context pressure
• + Free and open source, installable straight from GitHub

• − Claude Code only; no cross-agent support
• − No per-skill security scanning, license checks, or measured benchmarks
• − Curation reflects Composio's editorial view, which favors its own ecosystem tooling

https://github.com/composiohq/awesome-claude-plugins

A large open-source collection listing 345 Claude Code skills, agent skills, and plugins: 30+ agents, 70+ custom commands, 330+ skills, plus customizable references and scripts. What sets it apart is breadth of audience and platform. It explicitly targets Claude Code, Codex, Gemini CLI, Cursor, and 8 more coding agents, and its skills span far beyond engineering into marketing, product, compliance, C-level advisory, research, business operations, commercial and finance, and daily productivity. For a small-business owner who wants an agent to help with marketing or finance work rather than just code, that breadth is the draw. The trade-off is that it's a catalog, not a vetting service: there are no published trust scores, security scans, or sandbox benchmarks, so you're trusting the maintainer's packaging and your own testing. At 22k stars it has real community traction, which helps with maintenance, but with hundreds of skills the quality inevitably varies from excellent to filler.

pricing See pricing

• + Broadest non-engineering skill coverage of any collection here, including marketing, compliance, and finance
• + Cross-agent support covering Claude Code, Codex, Gemini CLI, Cursor, and more
• + Large community (22k stars) suggests active maintenance
• + Free and open source

• − No security scanning, license checks, or quality benchmarks published for individual skills
• − Volume over vetting: with 345 skills, quality varies considerably
• − No package manager or install tooling beyond cloning the collection

https://github.com/alirezarezvani/claude-skills

An open-source marketplace repo listing 425 plugins, 2,810 skills, and 200 agents for Claude Code, with a companion site at tonsofskills.com and its own CLI package manager called ccpi. It's the volume play in this comparison: no other collection here approaches 2,810 skills. The ccpi CLI is the practical differentiator, giving you a package-manager workflow for finding and installing pieces instead of hand-editing configs, which matters if you're setting up multiple agents or machines. Like the other community collections, it concentrates on Claude Code and publishes no per-skill security scanning, license verification, or sandbox benchmarks, so the burden of judging whether a specific skill is safe and effective falls on you. For a small business, that means it works best when you already know roughly what you need and want a fast way to install it, and less well when you're exploring and need signals about what actually works.

pricing See pricing

• + Largest catalog in this comparison by a wide margin
• + Ships its own CLI package manager (ccpi), rare among community collections
• + Open source and free

• − No published security, license, or quality vetting per skill
• − Claude Code focused only
• − Volume makes discovery harder without strong filtering

https://github.com/jeremylongshore/claude-code-plugins-plus-skills

A single hub for finding Claude Skills, Agents, Commands, Hooks, Plugins, and Marketplace collections that extend Claude Code, Claude Desktop, the Agent SDK, and OpenClaw. Think of it as a map of the ecosystem rather than a catalog of its own: it indexes the pieces scattered across GitHub so you don't have to search for each one. That breadth of surface area is its value, covering not just plugins but the full range of extension types, and reaching Claude Desktop and the Agent SDK, which most plugin marketplaces ignore. The limitation follows from the model: it's a discovery layer with curated links, so there's no per-skill verification, no security scanning, and no benchmarking. You still install and evaluate everything yourself. At 3.2k stars it's a smaller community project than the big collections, but for someone who wants one bookmark that covers the whole Claude extension ecosystem, it does that job cleanly.

pricing See pricing

• + Covers the widest range of extension types and Claude surfaces, including Desktop and the Agent SDK
• + Good starting point for exploring the ecosystem from one place
• + Free and open source

• − A discovery hub, not a vetted catalog: no security scans or benchmarks
• − Smaller community (3.2k stars) than the major collections
• − You still handle installation and evaluation yourself

https://github.com/davepoon/buildwithclaude