The 5 Best Claude Code Marketplace Alternatives in 2026

The Claude Code Marketplace is Anthropic's official, built-in system for discovering and installing plugins for Claude Code. Plugins are bundles that can include skills, agents, hooks, MCP servers, and other capabilities, and the marketplace lets you find and install them without building them yourself. You browse it by running /plugin inside Claude Code and using the Discover tab, and the official marketplace (claude-plugins-official) is automatically available when you start the tool. Third-party marketplaces can be added with /plugin marketplace add owner/repo-name. Before plugins existed, teams shared scattered CLAUDE.md files, hook scripts, and MCP configs; the marketplace made those pieces installable and shareable. The catch for small businesses: Claude Code itself has no free tier, and it keeps you inside Anthropic's models.

pricing 20/month

• + Zero-friction: it's inside Claude Code, no separate tool or account needed
• + Official Anthropic-managed directory of high quality plugins
• + Plugins package entire workflows (commands, MCP configs, hooks, project notes) into installable units
• + Supports third-party marketplace repos, so you can extend beyond the official directory

• − Requires a Claude Code subscription; no free tier, and Pro at $20/month gets tight with heavy agent use
• − Locked to Anthropic's models and a terminal-first workflow
• − Rate limits during peak hours, with heavy use pushing you to $100–$200/month Max plans
• − The official directory is smaller than community collections

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, license-checks, security-scans, and quality-judges each one, then benchmarks every skill in a sandbox with the same model, with and without the skill, publishing the Lift in percentage points. A trust score from 0 to 100 combines security, quality, license, maintenance, and community signals, and Lift is always shown separately from ratings. Per-agent loadouts let you toggle skills and tools per agent, and the platform tracks upstream changes, re-scans, re-benchmarks, and notifies you. Everything is free, funded by advertising that never touches scores.

pricing 0

• + Everything is free: the directory, the MCP connector, benchmarks, and alerts
• + The only option that security-scans, license-checks, and benchmarks skills with published Lift results
• + Trust score (0–100) from security, quality, license, maintenance, and community signals; money never touches scores
• + Works across Claude Code, Codex, Cursor, Gemini CLI, and chat-app agents
• + Per-agent loadouts mean each agent only sees the skills and tools you chose for it
• + Version pinning to commit SHA plus upstream change tracking and re-scanning

• − It's a directory and connector, not a plugin installer built into Claude Code, so installation is a separate step from your coding agent
• − As a newer independent platform, it doesn't have Anthropic's first-party integration with Claude Code
• − Benchmark Lift depends on its sandbox setup with the same model, which may not match your exact production workflow

https://skillzbench.com

Claude Plugins Official is Anthropic's own GitHub repository behind the Claude Code Marketplace, described as the official, Anthropic-managed directory of high quality Claude Code plugins. It's the repo the built-in marketplace reads from, and it's also usable directly: you can install from it with /plugin install github@claude-plugins-official, and if a plugin seems missing you refresh with /plugin marketplace update claude-plugins-official. The directory covers skills, agents, commands, hooks, and LSP servers, and Anthropic also maintains a separate Knowledge Work Plugins repo aimed at knowledge workers using Claude Cowork. With roughly 32k stars, it's one of the most-used collections in the ecosystem, though it's deliberately curated rather than exhaustive.

pricing 0

• + First-party curation from Anthropic, so quality is consistent
• + Free and open on GitHub
• + Same content the built-in marketplace serves, usable directly
• + Includes LSP plugins for TypeScript, Python, Rust, and Go, which improve every other workflow

• − Deliberately curated, so far fewer plugins than community mega-collections
• − Claude Code-focused; limited value for Codex, Cursor, or Gemini CLI users
• − No independent security scanning or benchmarking of plugin performance

https://github.com/anthropics/claude-plugins-official

Awesome Claude Plugins is Composio's curated open-source collection of Claude Code plugins, spanning slash commands, subagents, MCP servers, and hooks. It's one of the third-party marketplaces you can add to Claude Code with /plugin marketplace add, or install directly by cloning the repo and running the install command, which also sets up Composio Connect MCP so Claude Code can securely access apps like Jira, Linear, and GitHub from a single MCP. Composio's accompanying guide organizes plugins by type: code intelligence and LSP plugins (TypeScript, Pyright, Rust Analyzer, Go), MCP integration plugins (GitHub, Linear, Slack, Vercel, Sentry, Supabase), code review plugins, and security plugins like Security Guidance and Semgrep. One caveat Composio itself flags: installing many MCP servers at once can choke the LLM context window, which their dynamic tool loading is designed to mitigate.

pricing 0

• + Well-organized by plugin type, which makes choosing your first plugins easy
• + Includes Composio Connect MCP, which consolidates app access and avoids context-window bloat from many separate MCP servers
• + Free and open source
• + Covers high-value categories: LSPs, code review, and security scanning (Semgrep, Security Guidance)

• − Curated by a vendor whose own product (Composio) is part of the stack, so there's a commercial angle
• − Installing many MCP servers at once can choke the LLM context window, per Composio's own warning
• − No independent security scanning or benchmarking of listed plugins

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

Claude Skills by alirezarezvani is one of the largest community skill collections in the ecosystem: 345 Claude Code skills, agent skills, and plugins, including 30+ agents, 70+ custom commands, and 330+ skills with customizable references and scripts. What sets it apart from most Claude Code collections is breadth beyond engineering: it covers marketing, product, compliance, C-level advisory, research, business operations, commercial and finance skills, and general productivity. That makes it one of the few collections a non-developer small-business owner can actually mine for operational skills, not just code helpers. It works across Claude Code, Codex, Gemini CLI, Cursor, and 8 more coding agents, and it has earned roughly 22k stars on GitHub. The trade-off is the usual one for bulk community collections: no independent security scanning or benchmarking, so you're trusting the maintainer.

pricing 0

• + Unusual breadth: business, marketing, compliance, and finance skills, not just coding
• + Cross-agent support across 12+ coding agents
• + Free and popular (roughly 22k stars)
• + Customizable references and scripts

• − Bulk community collection with no independent security scanning
• − No benchmarking, so there's no evidence which of the 345 skills actually help
• − Quality varies across hundreds of community-contributed items

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

Claude Code Plugins Plus Skills, by jeremylongshore, is a volume-first open-source marketplace hosted at tonsofskills.com, listing 425 plugins, 2,810 skills, and 200 agents for Claude Code. It ships with its own CLI package manager, ccpi, so you can install items from the command line rather than copying files by hand, which puts it closer to a real marketplace experience than a plain awesome-list. The sheer size is the draw: if a niche workflow exists, there's a decent chance something here covers it. The flip side of 2,810 skills is discoverability and trust: the collection is community-assembled, with no independent security scanning, license checking, or benchmarking published, so a small business adopting from it takes on the vetting burden itself.

pricing 0

• + Largest raw selection on this list: 2,810 skills and 425 plugins
• + Includes a CLI package manager, so installs are scriptable
• + Free and open source
• + Good odds of finding niche or obscure workflows

• − No independent security scanning or benchmarking across the collection
• − Discoverability is hard at this scale without quality signals
• − Claude Code-centric
• − Community-maintained, so freshness and upkeep vary

https://tonsofskills.com

aitmpl.com is an index of Claude Code plugin collections, skill registries, and component marketplaces on GitHub, covering 34 collections. It's a meta-directory: rather than hosting plugins itself, it points you to the biggest repos with star counts and content-type tags (skills, agents, commands, hooks, MCPs, LSPs). Highlights include Ecc by affaan-m (229k stars), an agent harness performance optimization system spanning skills, instincts, memory, and security; Claude Mem by thedotmack (87k stars) for persistent context across sessions on Claude Code, Codex, Gemini, Copilot, OpenCode and more; Claude Hud by jarrodwatts (26k stars) for context and agent visibility; and HashiCorp's official Agent Skills collection. Star counts are the ranking signal here, which tells you popularity but nothing about security or measured performance. It's a useful starting map of the ecosystem, best paired with a verification layer.

pricing 0

• + One map of the whole ecosystem instead of hunting repos one by one
• + Content-type tags help you find, say, only MCP servers or only hooks
• + Free, with no install commitment
• + Includes vendor-backed collections (HashiCorp, Airtable, Timescale) alongside community ones

• − Popularity-ranked by stars, which says nothing about security or whether a skill actually helps
• − It's an index, not a marketplace: no installs, no versioning, no scanning
• − You still land on GitHub and vet everything yourself

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