The 5 Best Claude Code Marketplace Alternatives in 2026

The official, Anthropic-managed directory of Claude Code plugins. Plugins are packaged extensions that can bundle skills, agents, hooks, MCP servers, LSP integrations and slash commands, replacing the old scatter of CLAUDE.md files, hook scripts and custom configs teams used to share by hand. Because the marketplace ships with Claude Code itself, there's nothing to install or configure: you run /plugin, browse the Discover tab, and install with a single command such as /plugin install github@claude-plugins-official. Anthropic also publishes a separate Knowledge Work Plugins repository aimed at knowledge workers in Claude Cowork. The catalog leans toward developer workflows: LSP plugins for TypeScript, Python, Rust and Go, MCP integrations for GitHub, Linear, Slack, Vercel, Sentry and Supabase, plus code review and security plugins like Security Guidance and Semgrep. The trade-off is scope: it covers Claude Code only, and its curated list is deliberately smaller than the sprawling third-party collections, so niche or business-function skills usually live elsewhere.

pricing 0

• + Zero setup: the official marketplace is available automatically when you start Claude Code
• + Anthropic-managed curation gives it the strongest baseline trust of any list here
• + Plugins package skills, agents, hooks and MCP servers together, so installs are clean and reversible
• + Deep developer-focused picks like LSP plugins that give Claude jump-to-definition and type diagnostics

• − Claude Code only; no support for Codex, Cursor or Gemini CLI
• − Requires a paid Claude Code subscription underneath, and heavy agent use pushes toward $100-$200/month Max plans
• − Curated catalog is smaller than community marketplaces, so niche skills are often missing
• − No published per-skill benchmarking or Lift data

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. It exists because agent skills (SKILL.md instruction packages for Claude Code, Codex, Cursor, Gemini CLI and chat-app agents) have exploded across GitHub with no quality control: published research found roughly 26% of skills carry at least one vulnerability. SkillzBench runs five pillars, all live today. Discover continuously crawls and dedupes skills from GitHub. Verify pins every version to a commit SHA, then 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 so you can see what a skill actually adds. Equip gives you one free MCP server plus per-agent loadouts: toggle the skills and tools each agent gets, and the agent only sees what you chose. 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, with money never touching scores. Everything, including the directory, MCP connector, benchmarks and alerts, is free; revenue comes from blog advertising only.

pricing 0

• + Everything is free, including the MCP connector and benchmarks, with ad revenue that never touches scores
• + Only option here with per-skill security scanning, license checks and commit-SHA version pinning
• + Publishes measurable Lift data from sandboxed with/without-skill tests
• + Works across Claude Code, Codex, Cursor, Gemini CLI and chat-app agents, not just one tool
• + Per-agent loadouts prevent skills and tools from leaking between agents

• − It's a separate website and MCP server, not a command built into your coding agent
• − No curated Anthropic-style editorial picks; discovery is data-driven rather than hand-picked
• − Catalog size shifts as the crawler dedupes and re-verifies, so check the site for current numbers

https://skillzbench.com

Awesome Claude Plugins is the community-curated awesome-list for the Claude Code plugin ecosystem, maintained by Composio. It collects slash commands, subagents, MCP servers and hooks across categories, and it's the reference most roundups point to when the official marketplace doesn't have what you need. Installation is a one-liner from inside Claude Code: /plugin marketplace add owner/repo-name, or you can clone the GitHub repo directly and install from the local directory. Because it's an awesome-list rather than a managed product, its strength is breadth and community momentum: it surfaces plugins like Composio Connect for secure access to Jira, Linear and GitHub from a single MCP, plus the LSP, code review and security plugins that roundups rank highest. Its weakness is the flip side of that openness: entries are community-submitted, so there's no security scanning, license verification or benchmarking behind individual listings. If a plugin misbehaves, you find out in your own codebase.

pricing 0

• + Free and open source with strong community momentum
• + Broad category coverage beyond the official catalog
• + Simple install path that works with Claude Code's native marketplace command
• + Backed by Composio, which also ships dynamic tool loading to avoid choking the context window with multiple MCPs

• − No security scanning, license checks or benchmarks on listed plugins
• − Claude Code focused, so it won't serve Codex or Cursor workflows
• − Quality varies between community submissions

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

This collection packs 345 Claude Code skills, 30+ agents and 70+ custom commands into a single repository, aimed well beyond pure engineering. Categories span marketing, product, compliance, C-level advisory, research, business operations, commercial and finance, plus daily productivity, which makes it one of the few lists a small-business owner could actually mine for non-coding work. It's also notably cross-agent: the skills and plugins are built for Claude Code, Codex, Gemini CLI, Cursor and 8 more coding agents, so one install covers most of a modern agent stack. Everything is free and open source on GitHub, with customizable references and scripts included. The trade-off is accountability: this is a single-maintainer collection with no published security scanning, license verification or benchmark data, so the usual 26%-of-skills-have-vulnerabilities risk applies at full weight. It's best treated as a buffet to pick from deliberately, not a vetted catalog to install wholesale.

pricing 0

• + One of the largest single collections, with rare non-engineering categories like compliance and finance
• + Genuinely cross-agent, covering 12 coding agents
• + Free and open source with customizable references and scripts

• − Single maintainer, no published security scanning or verification
• − Volume makes it easy to install more than you need and bloat agent context
• − No benchmark data showing which skills actually help

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

Buildwithclaude is a single hub for finding Claude skills, agents, commands, hooks, plugins and marketplace collections, built by davepoon. Rather than being one more plugin pack, it's a discovery layer: it indexes the ecosystem so you can find the right collection before you install anything, covering Claude Code, Claude Desktop, the Agent SDK and OpenClaw. That makes it useful at the exact moment most people hit friction, when the official marketplace doesn't have a plugin and you don't know which third-party repo to trust. The hub format is its strength and its ceiling. It's free, it's current with the fast-moving ecosystem, and it saves real search time. But it doesn't verify what it indexes: no security scans, no license checks, no benchmarks, and no per-skill quality signals. You're still trusting whatever collection you land on. Pairing it with an independent verification layer is the sensible pattern if you use it as your front door.

pricing 0

• + Free and open source
• + Good entry point when the official catalog comes up short
• + Covers Claude Desktop and the Agent SDK, not just the CLI

• − Discovery only: no verification, scanning or benchmarking of what it indexes
• − Adds a layer between you and installation rather than shipping skills directly
• − Claude-ecosystem focused

https://github.com/davepoon/buildwithclaude

Claude Code Plugins Plus Skills, by jeremylongshore, is the volume play in this comparison: 425 plugins, 2,810 skills and 200 agents for Claude Code, published as an open-source marketplace at tonsofskills.com with its own ccpi CLI package manager. The ccpi tool matters because it turns browsing a giant list into an actual package-management workflow, closer to npm for skills than a static awesome-list. For teams that want to stock a wide library quickly, nothing else here approaches 2,810 skills. The caveats are the same ones that follow any high-volume, community-sourced catalog: no published security scanning, no license verification, and no benchmark data showing which of those thousands of skills actually lift performance. With independent research putting roughly 26% of agent skills in the vulnerable category, a catalog this size mathematically contains a meaningful number of risky entries. Treat it as a warehouse with a good forklift (ccpi), and verify before you load anything into production agents.

pricing 0

• + Largest raw catalog in this comparison by a wide margin
• + ccpi CLI gives real package-management workflow, not just a list
• + Free and open source

• − No published security scanning or quality verification at this scale
• − Claude Code only
• − Sheer volume makes selective, deliberate installation essential

https://tonsofskills.com