The 5 Best Claude Code Marketplace Alternatives in 2026

The Claude Code Marketplace is Anthropic's official directory for discovering and installing plugins, and it's available automatically when you start Claude Code. Anthropic describes plugins as bundles that can include skills, agents, hooks, MCP servers, LSP servers, and slash commands, so instead of sharing scattered CLAUDE.md files, hook scripts, and MCP configs, teams package those pieces and install them as needed. Installation runs through the /plugin command, with a Discover tab for browsing and direct installs from the official claude-plugins-official repo or third-party marketplaces like Awesome Claude Plugins. The ecosystem is broad: popular collections include the Anthropic-managed official directory (32k stars), Ecc (229k stars), Claude Mem for persistent context (87k stars), and Knowledge Work Plugins for non-engineering use. The marketplace itself is free; the cost sits in the Claude Code subscription underneath it, which has no free tier and runs $20/month on Pro, with heavy agent use pushing teams toward $100-$200/month Max plans.

pricing See pricing

• + Free to use and pre-configured inside Claude Code, no extra setup
• + Anthropic-managed official directory with quality-filtered plugins
• + Standardized packaging replaces scattered CLAUDE.md files, hook scripts, and MCP configs
• + Large third-party ecosystem, including collections with hundreds of skills and agents
• + Category coverage spans code intelligence, code review, security, and app integrations

• − Requires a Claude Code subscription, which starts at $20/month and has no free tier
• − Locked into Anthropic's models and a terminal-first workflow
• − Third-party marketplaces lack independent security scanning or benchmarking of skills
• − Loading many MCP servers at once can choke the LLM context window

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. Its premise is simple: skills (SKILL.md instruction packages for Claude Code, Codex, Cursor, Gemini CLI, and chat-app agents) are everywhere, but published research found roughly 26% of them carry at least one vulnerability, and nobody could tell which ones actually help. SkillzBench runs five pillars, all live today: Discover continuously crawls and dedupes skills from GitHub; Verify pins every version to a commit SHA and runs validation, license checks, security scans, and quality judgments; Benchmark tests every skill in a sandbox with the same model, with and without the skill, and publishes the Lift in percentage points; Equip provides 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 with re-scans, re-benchmarks, and notifications. A 0-100 trust score combines security, quality, license, maintenance, and community signals, and Lift is always shown separately. Everything is free, funded by blog advertising that never touches scores or listings.

pricing 0

• + Completely free, including the MCP connector, benchmarks, and alerts
• + Independent verification addresses the ~26% of skills found to carry at least one vulnerability
• + Works across Claude Code, Codex, Cursor, Gemini CLI, and chat-app agents rather than one ecosystem
• + Lift benchmarks show measurable impact, always displayed separately from ratings
• + Money never influences scores or listings; revenue comes from blog ads only

• − It's a skills directory and connector, not a coding environment or IDE
• − Trust scores and Lift results depend on what has been crawled and benchmarked so far
• − Focused on skills, so it doesn't replace plugin bundles that include hooks and commands

https://skillzbench.com

Cursor is an AI-native code editor forked from VS Code and the closest drop-in replacement if you want Claude Code's agentic coding inside a graphical IDE. Cursor 3 rebuilt the interface around agents rather than files, with an Agents Window to launch, monitor, and review multiple agents across repos from one place. It supports Claude, GPT, Gemini, Grok, and DeepSeek, plus its built-in Composer model: Composer 2.5 ties Claude Opus 4.7 on SWE-Bench Multilingual (79.8% vs 80.5%) and Terminal-Bench 2.0 (69.3% vs 69.4%) while running at roughly a tenth of the per-token cost on its standard tier, though the Fast tier used by default costs $3/$15 per million tokens. Background Agents spawn up to 8 parallel agents in isolated cloud VMs that clone your repo, run tests, and open merge-ready PRs. Automations trigger on GitHub events, Slack messages, Linear changes, or cron schedules, and BugBot Autofix runs on every PR, with over 35% of proposed fixes merged directly. The trade-off is the credit system: heavy Pro users report actual spend of $40-$50/month once overages hit, and there's no precise real-time meter.

pricing 20/month · free trial

• + Visual IDE with the gentlest learning curve among agentic coding tools
• + Multi-model flexibility including Anthropic, OpenAI, Google, and its own Composer model
• + Parallel agents on isolated git worktrees with checkpoints for rollback
• + Cloud agents and event-driven automations for hands-off workflows

• − Credit-based pricing makes cost planning fiddly; heavy Pro users report $40-$50/month actual spend
• − Trails Claude Code in developer sentiment: 19% vs 46% 'most loved' in Pragmatic Engineer's early 2026 survey
• − No precise real-time meter for credit consumption

https://cursor.com

Codex is OpenAI's coding agent, available as a desktop app for 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 the primary experience: a command center for managing multiple agents at once, with parallel agents running on isolated Git worktrees so you can explore different approaches conflict-free. Skills extend Codex beyond code generation, bundling instructions and scripts to 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 schedules for daily issue triage, CI failure summaries, or release briefs, with results landing in a review queue. GitHub delegation is a standout: tag @codex on a PR and a cloud agent starts working, and the openai/codex-action GitHub Action handles CI/CD workflows including waiting for an approved-plan label before executing. The CLI includes OS-level kernel sandboxing that physically prevents file access outside your project or unauthorized network calls. It's bundled with ChatGPT plans with no separate billing, but it runs OpenAI models only.

pricing 20/month

• + Bundled with ChatGPT plans you may already pay for, no separate billing
• + Over 2 million weekly developers, strong hand-off-and-review workflow
• + Open-source CLI with kernel-level sandboxing
• + Lower token use per task than Claude Code per DataCamp's comparison

• − OpenAI models only, no Anthropic or Google model access
• − Desktop app is macOS and Windows only
• − Skills ecosystem is younger than Claude Code's plugin marketplace

https://openai.com/codex

OpenCode is the open-source (MIT) terminal CLI with the most model flexibility of any tool in this comparison: 75+ providers, zero markup on model costs, and 120,000+ GitHub stars. Its headline feature is subscription piggybacking: run /connect and authenticate with accounts you already pay for, including GitHub Copilot ($10-$39/month) or ChatGPT Plus ($20/month), and OpenCode routes agent work through them instead of billing you separately. That makes it the cheapest serious option if you already hold any AI subscription, and completely free if you run local models, since it's model-agnostic and supports local LLMs for privacy-sensitive work. Builder.io lists it as the pick for model flexibility, subscription piggybacking, and zero-markup pricing, and DataCamp recommends it specifically for privacy and local model use. The trade-offs are what you'd expect from a terminal-first open-source project: no visual editor, no built-in cloud agent fleet, and you're responsible for your own setup and updates. For small businesses with a developer who lives in the terminal, though, it's the strongest cost-to-capability ratio here.

pricing 0

• + Free and open source, no vendor lock-in
• + Reuses subscriptions you already pay for instead of adding a new bill
• + Broadest model flexibility in the group, including local models
• + Large, active open-source community

• − Terminal-only, no visual editor or diffs
• − No managed cloud agents or automations
• − Self-managed setup and maintenance

https://opencode.ai

GitHub Copilot is the pragmatic pick for teams that already live on GitHub, and DataCamp calls it hard to beat at $10/month for that crowd. It pairs inline autocomplete with native GitHub.com integration, so code suggestions, agent work, and pull-request workflows sit inside the platform your team already uses rather than in a separate tool. A free tier exists for individual developers, and paid plans run $10-$39 per user per month, the cheapest paid entry among the full coding agents here. Copilot isn't trying to match Cursor's parallel agent fleets or Codex's desktop command center; its differentiator is friction removal for GitHub-centric teams: issues, PRs, and reviews all stay in one place, and onboarding is nearly zero because most developers already have GitHub accounts. The agentic capabilities are more limited than the dedicated agent platforms, and model choice is narrower than Cursor or OpenCode. For a small business with a couple of developers shipping through GitHub, it's often the lowest-cost way to get meaningful AI assistance without changing anyone's workflow.

pricing 10/month · free trial

• + Cheapest paid entry among full coding agents at $10/user/month
• + Free tier for individuals
• + Zero workflow change for GitHub-centric teams
• + Backed by GitHub and Microsoft infrastructure

• − Fewer agentic features than Cursor, Codex, or Claude Code
• − Narrower model choice than Cursor or OpenCode
• − Value depends heavily on already using GitHub

https://github.com/features/copilot