The 5 Best Claude Code Marketplace Alternatives in 2026

The Claude Code Marketplace is Anthropic's official, built-in directory for Claude Code plugins. Anthropic describes plugins as bundles that can include skills, agents, hooks, MCP servers, and other capabilities, and the marketplace lets you discover and install them without building them yourself. It ships automatically with Claude Code: run /plugin, browse the Discover tab, or install directly with /plugin install. The official repo, claude-plugins-official, is an Anthropic-managed directory of what it calls high quality plugins, covering LSPs, code review, security guidance, and MCP integrations for services like GitHub, Linear, Slack, Vercel, Sentry, and Supabase. Third-party marketplaces can be added too, via /plugin marketplace add. It's the most frictionless option if you live inside Claude Code, but it's Claude Code only, and the directory doesn't publish per-plugin security scans or independent benchmark results.

pricing 0

• + Zero setup: the official marketplace is available automatically when you start Claude Code.
• + One package format covers skills, agents, hooks, MCP servers, and LSPs.
• + Anthropic-managed official directory, plus an ecosystem of community marketplaces you can add.
• + Strong plugin coverage for code intelligence, code review, security, and MCP integrations.

• − Claude Code only; skills and plugins don't carry over to Codex, Cursor, or Gemini CLI.
• − No free tier for Claude Code itself, and heavy use pushes you to $100–$200/mo Max plans.
• − No published per-plugin security scans or independent benchmark results.
• − Installing many MCP plugins at once can choke the LLM context window, per Composio's guide.

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) are everywhere, but nobody can tell which ones actually help and which are risky: published research found roughly 26% of skills carry at least one vulnerability. SkillzBench continuously crawls and dedupes skills from GitHub, pins every version to a commit SHA, validates and license-checks each one, security-scans it, and quality-judges it. It then benchmarks every skill in a sandbox with the same model, with and without the skill, and publishes the Lift in percentage points. A 0–100 trust score combines security, quality, license, maintenance, and community signals, and money never touches scores. Per-agent loadouts let you toggle skills and tools per agent so each agent only sees what you chose, and the platform tracks upstream changes, re-scans, re-benchmarks, and notifies you.

pricing 0

• + Everything is free: the directory, the MCP connector, benchmarks, and alerts.
• + The only option here that publishes independent security scans and measured Lift for every skill.
• + Works across Claude Code, Codex, Cursor, Gemini CLI, and chat-app agents, not just one tool.
• + Per-agent loadouts stop context bloat by showing each agent only what you selected.
• + Commit-SHA pinning plus upstream tracking means you know when a skill changes and gets re-verified.

• − It's a skills directory and MCP connector, not a coding agent or IDE, so it complements rather than replaces your editor.
• − As an independent directory, it doesn't give you the official Anthropic plugin bundles like hooks and LSPs packaged together.
• − Revenue comes from advertising around blog articles (though ads never influence scores or listings).

https://skillzbench.com

Cursor is the market leader in AI-powered editors, a VS Code fork with AI woven through every layer, reporting $2B ARR and over 1 million paying users. Where Claude Code is terminal-first, Cursor gives you a visual agent workspace: an Agents Window to launch, monitor, and review multiple agents across repos. It supports Claude, GPT, Gemini, Grok, and DeepSeek, plus its own Composer model built for agentic coding. Cursor 3's Composer 2.5 ties Claude Opus 4.7 on SWE-Bench Multilingual (79.8% vs 80.5%) at roughly a tenth of the per-token cost on the standard tier. Background Agents run up to 8 parallel agents in isolated cloud VMs that open PRs when done, and Automations trigger agents from GitHub events, Slack messages, or schedules. BugBot Autofix runs on every PR, with over 35% of proposed fixes merged directly. In Pragmatic Engineer's early 2026 survey, Claude Code took 46% 'most loved' against Cursor's 19%, so the visual workspace trades some developer affection for a gentler learning curve.

pricing 20/month

• + Visual IDE with a gentler learning curve than terminal-first agents.
• + Model flexibility across five providers plus its own Composer model.
• + Parallel background agents, worktrees, and event-driven automations.
• + Free Hobby tier with no credit card required.

• − Credit system makes cost planning fiddly; Pro users report $40–$50/mo actual spend once overages kick in.
• − Trails Claude Code in developer sentiment (19% vs 46% 'most loved' in Pragmatic Engineer's early 2026 survey).
• − Not open source.

https://cursor.com

Codex is OpenAI's coding agent, available as a desktop app (macOS and Windows), an open-source CLI (Apache 2.0), and a VS Code extension, with over two million developers using it weekly. The desktop app is a command center for managing multiple agents at once: parallel agents with Git worktrees run on isolated copies of your repo so you can explore approaches simultaneously. Skills extend Codex beyond code generation, bundling instructions, resources, and scripts so agents can connect to Figma, deploy to Vercel or Netlify, manage Linear issues, or generate images, and they work across the app, CLI, and IDE extension. Automations run agents on a schedule for issue triage, CI failure summaries, or release briefs, with results landing in a review queue. GitHub works as a task interface: tag @codex on a PR and a cloud agent starts working. The open-source CLI adds OS-level kernel sandboxing that physically prevents file access outside your project. It's bundled with ChatGPT plans, so there's no separate billing, but it only runs OpenAI models.

pricing 20/month

• + No separate billing if you already pay for ChatGPT Plus, Pro, or Business.
• + Skills system mirrors Claude Code's, and works across app, CLI, and IDE.
• + Open-source CLI with real kernel-level sandboxing.
• + Over two million weekly developers and deep GitHub integration.

• − OpenAI models only; no Claude or Gemini.
• − Lower token use per task, but you're locked into OpenAI's ecosystem.
• − No independent verification or benchmarking of community skills.

https://openai.com/codex

OpenCode is the open-source (MIT) CLI with the most model flexibility on this list: 75+ providers, zero markup, and the ability to reuse subscriptions you already pay for, with 120,000+ GitHub stars. Its headline feature is subscription piggybacking: run /connect, authenticate with GitHub Copilot ($10–$39/mo), ChatGPT Plus ($20/mo), or other existing subscriptions, and use those models inside OpenCode without new API bills. It's model-agnostic and can run local LLMs, which makes it the privacy pick for teams that don't want code leaving their infrastructure. As a terminal-first CLI it's aimed squarely at developers, so it lacks the visual editing of Cursor or the desktop command-center feel of Codex, but for anyone who wants a free, flexible, no-lock-in agent harness it's hard to beat. It doesn't ship its own verified skills marketplace, which is where an independent directory like SkillzBench fills the gap.

pricing 0

• + Free and open source with zero markup on model costs.
• + Reuse subscriptions you already pay for instead of new API bills.
• + Runs local models for privacy-sensitive work.
• + Huge community (120,000+ GitHub stars).

• − Terminal-first; no visual editor for non-developers.
• − No built-in skills marketplace or verification layer.
• − You manage model setup and costs yourself.

https://opencode.ai

Builder.io is a multiplayer AI development platform where entire teams build together on your real codebase: engineers, designers, PMs, and stakeholders work in the same environment, with every change landing as a reviewable PR. Describe a feature and Builder spins up agents in cloud containers; a designer can jump into a live visual preview for pixel-level adjustments, and a PM can prompt changes directly in context. It runs 20+ parallel agents simultaneously, each in its own cloud container, so you can assign a batch of tickets and come back to finished pull requests. Slack and Jira work as entry points: tag @Builder.io in Slack and it reads context, creates a branch, builds the feature, and drops preview links in the thread. Review agents perform static analysis and flag security vulnerabilities before a human looks at the PR, and a Figma-like editor enforces your design system while Figma-to-Code converts designs into React, Vue, Angular, or Vue using your components and tokens. MCP servers connect Supabase, Netlify, Zapier, and custom tools at the workspace level.

pricing See pricing

• + Non-engineers (designers, PMs) can contribute directly to real code.
• + Massive parallelism: 20+ agents at once on batched tickets.
• + Built-in review agents catch security issues before human review.
• + Every change lands as a reviewable PR.

• − It's a whole development platform, a much bigger commitment than a plugin marketplace.
• − No pricing details published in the sources reviewed, so budgeting takes a sales conversation.
• − Overkill if you just want to extend an existing coding agent with skills.

https://builder.io