657 boosters for "Claude" — open source, verified from GitHub, ready to install
"name": "spanish-research-grants", "name": "Dr. Pablo Zulet Fraile", "email": "pablozuletfraile@gmail.com"
"name": "claude-at-proto-plugins", "name": "John Whiteside" "description": "Claude Code plugins for AT Protocol and Bluesky",
An MCP server that discovers and exposes Agent Skills for Claude, enabling developers to extend AI capabilities with reusable skill modules. Useful for AI engineers building skill-based agent systems.
"name": "init-workspace", "description": "Initialize a multi-agent workspace with shared configuration for Claude Code, Codex CLI, and Gemini CLI", "homepage": "https://github.com/xiaolai/init-workspace",
A refactoring task booster that helps developers improve code quality by optimizing structure, naming, performance, and maintainability with before/after comparisons. Ideal for developers seeking to enhance existing codebases and establish better coding practices.
"name": "claude-memory-plugin", "description": "Persistent knowledge storage with semantic search, graph linking, and contextual gotcha injection", "name": "Gareth Williams",
Analyzes Claude Code conversation history to identify friction patterns and suggest systematic improvements to skills and workflows. Useful for developers who want to review multi-session interactions and iteratively refine their AI assistant usage.
A skill that enables Claude Code to query other LLMs (ChatGPT, Gemini, DeepSeek, Grok) through MCP servers by routing requests based on the first word of a query. Useful for developers who need multi-LLM comparison or fallback capabilities within Claude Code environments.
"name": "Jonathan Arfa" "name": "critique-loop", "source": "./plugins/critique-loop",
A collection of 48 specialized AI subagents for Claude Code that automate domain-specific development tasks like backend architecture, frontend development, and DevOps. Developers benefit from having expert agents pre-configured for different roles, reducing context-switching and improving productivity across the full software development lifecycle.
"name": "legacybridge-cc-plugins", "version": "1.17.0", "description": "Collection of Claude Code plugins and tools by Frank Wang",
Agent Chatroom MCP Server enables real-time communication and coordination between parallel Claude Code agents, allowing developers to build collaborative multi-agent systems with synchronized messaging and task distribution.
"$schema": "https://anthropic.com/claude-code/marketplace.schema.json", "name": "windows-pc-optimizer", "description": "Windows PC optimization toolkit for Claude Code",
"name": "forge-council", "description": "Multi-agent council orchestration via Claude Code agent teams.", "author": { "name": "Martin Zeman" },
This booster generates comprehensive documentation for code, covering overview, API details, setup, usage guides, and technical specifications in well-formatted markdown. Developers working with Claude Code can use it to automate documentation generation and ensure consistent, thorough project documentation.
A skill booster that enables Claude Code to query the Exa search API through the MCP server, allowing developers to integrate web search capabilities into their coding workflows.
This booster enables Claude to query documentation references via the ref MCP server within Claude Code. It's designed for developers who need quick access to reference materials while coding.
"name": "Chris von Csefalvay", "email": "chris@chrisvoncsefalvay.com" "description": "Claude Code skills marketplace for fine-tuning language models with Unsloth",
"name": "claude-local-docs", "version": "1.0.23", "description": "Local-first Context7 alternative — indexes JS/TS dependency docs with a 4-stage RAG pipeline (vector + BM25 + RRF + cross-encoder reranking). Uses TEI Docker containers for GPU-accelerated embeddings and reranking.",
An MCP server that integrates Obsidian vaults as a knowledge management layer for AI assistants like Claude and Copilot, enabling developers and knowledge workers to leverage their personal notes and documentation in AI-powered workflows.
skill-auditor is an expert reviewer that evaluates SKILL.md files against Claude Code Skills best practices, helping developers ensure their skills meet structural and effectiveness standards. It's essential for skill creators and maintainers who want to validate compliance before publishing.
claude-ultrathink enforces single-task linear processing to prevent parallel execution and ensure focused sequential task completion. Developers using claude_code benefit by maintaining task discipline and reducing context-switching overhead.
UltraThink activates a structured single-agent development methodology using linear task execution and TodoWrite planning, designed for developers building Claude-based agents who need organized, step-by-step task management.
This MCP server transforms linear AI reasoning into structured, auditable thought graphs using DAG (directed acyclic graph) representations, enabling developers to visualize and verify AI decision-making processes with full traceability.
