The 5 Best Claude Code Marketplace Alternatives in 2026

Anthropic's official plugin directory for Claude Code. Plugins are bundles that can include skills, agents, hooks, MCP servers and other capabilities, and the marketplace lets you discover and install them without building anything yourself. It's available automatically when you start Claude Code: run /plugin, browse the Discover tab, and install directly from the claude-plugins-official repo. Third-party marketplaces can be added with /plugin marketplace add owner/repo-name. The official Anthropic-managed collection (Claude Plugins Official) sits at roughly 32k GitHub stars, and Anthropic also publishes a separate Knowledge Work Plugins repo aimed at non-engineering users. The strength is friction-free installation and Anthropic's own quality bar on the official listings. The weakness is that discovery and trust are shallow: you see what a plugin claims, not whether it's been security-scanned or benchmarked, and published research indicates around 26% of agent skills carry at least one vulnerability.

pricing See pricing

• + Zero setup: the official marketplace is available the moment you start Claude Code
• + Installs packaged workflows (skills, hooks, MCP configs, commands) in one command
• + Anthropic-managed official directory with a stated quality bar
• + Easy to extend with third-party marketplaces from any GitHub repo

• − No per-skill security scanning or trust scores on listings
• − No benchmarking, so you can't see whether a skill actually improves results
• − Tied to the Claude Code ecosystem; skills don't transfer cleanly to other agents

https://code.claude.com/docs/en/plugins

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. It exists because agent skills (SKILL.md instruction packages for Claude Code, Codex, Cursor, Gemini CLI and chat-app agents) have multiplied faster than anyone's ability to vet them, and published research found roughly 26% of skills carry at least one vulnerability. The platform runs five pillars today: Discover continuously crawls and dedupes skills from GitHub; Verify pins every version to a commit SHA and validates, license-checks, security-scans and quality-judges it; 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 the agent only sees the skills and tools you toggled on; and Stay current tracks upstream changes, re-scans, re-benchmarks and notifies you. A 0-100 trust score combines security, quality, license, maintenance and community signals, and Lift is always shown separately. Advertising in blog articles funds the platform; money never touches scores.

pricing 0

• + Everything is free: directory, MCP connector, benchmarks and alerts
• + Security scanning addresses the ~26% vulnerability rate found in agent skills research
• + Lift benchmarks show measured impact, not vendor claims
• + Works across Claude Code, Codex, Cursor, Gemini CLI and chat-app agents
• + Trust scores are independent; ad revenue never influences them

• − It's a verification and connection layer, not an in-editor plugin browser, so it adds a step beyond /plugin
• − You won't find specific skill counts or Lift figures here; check skillzbench.com for current numbers

https://skillzbench.com

Composio's curated GitHub collection of Claude Code plugins, organized by what the plugin does rather than by popularity alone. It's the reference most developers hit when they want a recommendation: code intelligence and LSP plugins (TypeScript LSP, Pyright, Rust Analyzer, Go LSP), 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 collection doubles as a Composio on-ramp: cloning it installs Composio Connect MCP, which lets Claude Code securely access apps like Jira, Linear and GitHub from a single MCP server, with dynamic tool loading to avoid choking the context window when multiple MCPs are installed. Installation is a single /plugin marketplace add command or a git clone. Curation is opinionated and genuinely useful, but it stops at recommendations: there's no per-skill security scan, version pinning or benchmark data.

pricing See pricing

• + Best-in-category recommendations with reasoning (e.g. TypeScript LSP first for typed codebases)
• + Flags a real problem: stacking multiple MCP servers floods the LLM context window
• + Free and open source

• − Curation, not verification: no security scans or benchmarks per plugin
• − Doubles as a funnel into Composio's commercial platform

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

A directory that indexes the most popular Claude Code plugin collections, skill registries and component marketplaces on GitHub: 34 collections at last count, each tagged by what it contains (skills, agents, commands, hooks, MCPs, LSPs) and ranked by GitHub stars. The heavyweights it surfaces include Ecc by affaan-m (229k stars, an agent harness with skills, memory and security), Claude Mem (87k stars, persistent context across sessions), Claude Plugins Official (32k stars), Compound Engineering Plugin (23k stars), Claude Skills by alirezarezvani (22k stars, 345 skills), and smaller specialized entries like Playwright Skill for browser automation and Pg Aiguide for Postgres. It's the fastest way to see the whole third-party ecosystem at a glance and jump straight to a repo. What it doesn't do is evaluate: star counts measure popularity, not safety, and nothing on the page is security-scanned or benchmarked.

pricing See pricing

• + Broadest single-page view of the third-party marketplace ecosystem
• + Tags make it easy to filter by what you actually need
• + Free to browse

• − Popularity-ranked only; no quality or security evaluation
• − You leave the site immediately to install, so there's no ongoing tooling

https://www.aitmpl.com/plugins/

One of the largest single skill collections in the ecosystem: 345 Claude Code skills and agent skills, including 30+ agents, 70+ custom commands, 330+ skills, plus customizable references and scripts. Its differentiator is audience: alongside engineering content it covers marketing, product, compliance, C-level advisory, research, business operations, and commercial and finance work, which makes it unusually relevant to small-business owners who want agents doing non-engineering work. Compatibility is also broad: the repo targets Claude Code, Codex, Gemini CLI, Cursor and eight more coding agents. The trade-off is the usual one for mega-repos: breadth without verification. Nothing in the collection is individually security-scanned or benchmarked, and with 345 items the quality bar inevitably varies item to item. Given research showing roughly 26% of agent skills carry at least one vulnerability, treating this as a browsing pool rather than a trusted source is the sensible posture.

pricing See pricing

• + Rare coverage of non-engineering, business-oriented skills
• + Huge breadth across 13+ coding agents
• + Free and open source

• − No per-skill verification, scanning or benchmarking
• − Volume makes quality uneven; you're on your own picking winners

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

A curated Claude Code plugin marketplace from the developer known as obra, listing commands, MCP servers and skills. It appears in the aitmpl index with around 1.1k stars, which puts it in the focused-and-maintained tier rather than the mega-repo tier. The appeal is editorial: a small, curated set of plugins chosen by one maintainer, installable through the standard /plugin marketplace add flow, is easier to reason about than a 345-item dump. It's a good second marketplace to add once you've exhausted the official directory. The limitations mirror every GitHub-listed marketplace: no published security pipeline, no version pinning to commit SHAs, no benchmark data on whether the plugins actually lift agent performance, and no cross-agent story if your team also runs Codex or Cursor.

pricing See pricing

• + Small, opinionated curation that's easy to evaluate
• + Free and open source

• − Much smaller selection than the official directory or mega-repos
• − No verification, benchmarking or cross-agent support

https://github.com/obra/superpowers-marketplace