The 5 Best Claude Code Marketplace Alternatives in 2026

The Claude Code Marketplace is Anthropic's official plugin system for Claude Code. Anthropic describes plugins as bundles that can include skills, agents, hooks, MCP servers, and other capabilities, and the marketplace lets you discover and install those extensions without building them yourself. Before plugins existed, teams shared scattered CLAUDE.md files, hook scripts, MCP configs, and custom slash commands; the marketplace packages those pieces into installable units. The official marketplace (claude-plugins-official) is automatically available when you start Claude Code, and you browse it with the /plugin command's Discover tab or install directly with /plugin install. Anthropic also maintains separate collections like Knowledge Work Plugins for knowledge workers. The trade-off is curation without verification depth: the directory is Anthropic-managed and quality-focused, but there is no published per-plugin security scan, license check, or sandboxed benchmark showing how much a plugin actually improves your agent's output. You are also tied to the Claude Code ecosystem, and Claude Code itself has no free tier, with Pro at $20/month and heavy use pushing toward $100-$200/month Max plans.

pricing 1 tiers

• + Zero setup: the official marketplace is available automatically when you start Claude Code
• + Anthropic-managed directory focused on high-quality plugins
• + Plugin packaging replaces scattered CLAUDE.md files, hook scripts, and MCP configs
• + Supports third-party marketplaces if you want to go beyond the official catalog

• − No published security scans, license checks, or benchmarks for individual plugins
• − Locks you into Anthropic's ecosystem and Claude Code's terminal-first workflow
• − Claude Code has no free tier; heavy agent use pushes toward $100-$200/month Max plans
• − Installing many MCP-heavy plugins can choke the LLM context window

https://code.claude.com/docs/en/discover-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 exploded in number while nobody could tell which ones actually help and which are risky; 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 each one; 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 Stay current tracks upstream changes, re-scans, re-benchmarks, and notifies you. Each skill gets a Trust Score from 0-100 built from security, quality, license, maintenance, and community signals, and Lift is always shown separately from ratings. Everything is free, funded by blog advertising that never influences scores or listings.

pricing 0/month

• + Completely free, including the MCP connector, benchmarks, and alerts
• + Addresses the documented risk that about 26% of agent skills carry at least one vulnerability
• + Works across Claude Code, Codex, Cursor, Gemini CLI, and chat-app agents, not just one ecosystem
• + Lift benchmarks give you measured evidence of whether a skill actually helps
• + Money never touches scores; ad revenue is kept separate from ratings and listings

• − It is a directory and connector, not a full plugin package manager inside Claude Code's native /plugin flow
• − Benchmark Lift figures vary by task, so a high Lift on one task type does not guarantee the same gain on yours
• − Newer platform, so community recognition trails the official Anthropic directory

https://skillzbench.com

Awesome Claude Plugins is Composio's open-source, curated collection of Claude Code plugins, installable as a third-party marketplace with /plugin marketplace add composiohq/awesome-claude-plugins. Its standout piece is the Composio Connect MCP, which lets Claude Code securely access apps like Jira, Linear, GitHub, and more from a single MCP server. That matters because the research notes a real problem with stacking individual MCP plugins: install Linear, GitHub, Supabase, and Vercel separately and you choke the LLM context window as each server loads all its tools. Composio addresses this with dynamic tool loading (tools load as needed based on context), a remote workbench for chaining multiple tools into complex workflows, and a bash tool for handling large files outside the context window. Setup is quick, with Composio claiming you can get started within a minute at its dashboard. The collection is curated rather than exhaustive, and like other community marketplaces it does not publish per-plugin security scans or sandbox benchmarks, so you are trusting the curation rather than independent verification.

pricing See pricing

• + Solves the multi-MCP context-window problem with dynamic tool loading
• + One MCP connection covers many business apps instead of installing each separately
• + Curated list maintained by an active developer-tools company
• + Fast setup, with Composio claiming a minute or less to get started

• − No independent security scanning or benchmarking of the plugins it lists
• − Requires manually adding the marketplace repo in Claude Code
• − Curated scope means fewer plugins than the largest community collections

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

Superpowers Marketplace by obra is a curated Claude Code plugin marketplace listed on the aitmpl plugin collections directory with about 1.1k GitHub stars. It bundles commands, MCP servers, and skills into a single curated collection, in the spirit of a hand-picked toolbox rather than a sprawling index. The maintainer-driven curation is its main appeal: everything in the collection has been chosen to work together, which reduces the guesswork that comes with installing random plugins from unvetted sources. It installs as a third-party marketplace inside Claude Code using the standard /plugin marketplace add flow, so existing Claude Code users need no new tooling. The limitations mirror those of most community collections: curation is a human judgment call, not a verification process, so there are no published security scans, license checks, or sandbox benchmarks behind each entry, and the collection is small compared with giants like the alirezarezvani Claude Skills repo (345 skills) or Ecc (229k stars). If a plugin in the collection stops being maintained, nothing automatically re-scans it or tells you.

pricing See pricing

• + Hand-curated, so entries are chosen to work well together
• + Covers commands, MCPs, and skills in one marketplace
• + Free and open to any Claude Code user

• − Small collection (about 1.1k stars) compared with the largest community repos
• − No published security scans, license checks, or benchmarks
• − No automated re-scanning or notifications when upstream plugins change

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

Buildwithclaude by davepoon is a single hub for finding Claude Skills, Agents, Commands, Hooks, Plugins, and Marketplace collections, listed on the aitmpl directory with about 3.2k GitHub stars. Its differentiator is scope: rather than serving only Claude Code, it covers Claude Code, Claude Desktop, the Agent SDK, and OpenClaw, which makes it useful if your team touches several Claude surfaces and wants one place to look. The hub organizes resources by type (skills, agents, commands, hooks, plugins), so you can browse what exists before committing to an install. As an aggregator, its value depends on the underlying repos it points to, and it does not perform its own verification: there are no security scans, license checks, or sandbox benchmarks published for the resources it indexes, and no change-tracking to alert you when an upstream skill is updated or breaks. For a small business, that means Buildwithclaude is a good discovery starting point, but you still carry the risk assessment yourself unless you pair it with a verification layer.

pricing See pricing

• + Broad coverage across Claude products, not just the CLI
• + Organized by resource type, which makes browsing straightforward
• + Free and open source

• − Aggregator only: no security scanning, license checks, or benchmarks of its own
• − No upstream change tracking or alerts
• − Modest collection size relative to the biggest community repos

https://github.com/davepoon/buildwithclaude

Claude Skills by alirezarezvani is one of the largest community skill collections, listed on the aitmpl directory with about 22k GitHub stars. It packs 345 Claude Code skills and agent skills, 30+ agents, 70+ custom commands, and 330+ skills with customizable references and scripts, and it works with Claude Code, Codex, Gemini CLI, Cursor, and 8 more coding agents. Coverage is unusually broad for a community repo: engineering, marketing, product, compliance, C-level advisory, research, business operations, commercial and finance, and daily productivity. For a small business, that breadth is the draw: a marketing manager and a developer can both find relevant skills in the same collection, and the multi-agent support means skills are not locked to one tool. The trade-off is volume without verification. With hundreds of skills in one repo, there is no published security scanning, license checking, or sandbox benchmarking, and the published research finding that roughly 26% of agent skills carry at least one vulnerability applies to community collections like this one. You also get no change alerts when upstream skills are updated.

pricing See pricing

• + Enormous breadth of skills across business and technical functions
• + Works with many coding agents, not just Claude Code
• + Large community footprint (about 22k stars) suggests active use

• − No independent security scanning or benchmarking despite the volume
• − Hundreds of skills make quality uneven and discovery time-consuming
• − No change tracking or alerts when skills are updated upstream

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