Claude MCP Tools

This repository has been archived.All functionality has been superseded by OpenClaw’s native skill ecosystem, Desktop Commander, and community MCP connectors.

Routing concept (AgenticSeek): The smart provider-routing idea lives on as a design reference for future OpenClaw cost-aware routing features.

Decision date: 2026-02-14Reason: 14 MCP servers evaluated; none provide unique value beyond existing tooling.

Professional Model Context Protocol (MCP) server ecosystem for Anthropic's Claude Desktop and Claude Code CLI integration.

🎯Current Status: Fully Tested Production-Ready Multi-Provider AI System✅

- Latest Update: June 2, 2025
- Testing Status: ⭐COMPREHENSIVE TESTING COMPLETE- 94/100 overall score
- Server Count: 21 operational MCP servers (consolidated from 25)
- Tool Validation: All 7 core MCP tool categories verified working perfectly
- Performance: Sub-second response times across all tools
- Integration: 3/3 complex workflow scenarios tested and functional
- System Status: Claude Code Integration tested with 100% success rate
- Achievement: Comprehensive memory system and enhanced context retention

This repository provides a production-ready ecosystem of 21 MCP servers enabling advanced AI-assisted development workflows. The servers facilitate Claude's interaction with external systems, APIs, and development tools through standardized Model Context Protocol implementations.

- Hybrid AI Development: Seamless integration between Claude Desktop and Claude Code CLI
- Multi-Provider AI Routing: Smart routing between local and cloud AI providers with cost optimization
- Advanced Memory Management: Persistent context storage with rich entity-relation graphs
- Desktop Automation: Comprehensive GUI automation and testing capabilities
- Development Workflow Enhancement: Code analysis, formatting, security scanning, and project management
- Claude Code Integration MCP⭐TESTED & VERIFIED- Task delegation and project analysis
- AgenticSeek MCP- Multi-provider AI routing (Local DeepSeek, OpenAI, Google Gemini)
- Vibetest MCP- Multi-agent browser QA testing swarm
- Code Formatter MCP- Black/Prettier wrapper for code formatting
- Security Scanner MCP- Vulnerability scanning with pip-audit and npm audit
- Windows Computer Use MCP- Desktop automation with screen capture
- Containerized Computer Use MCP- Docker-isolated GUI automation
- API Gateway MCP- Unified routing for OpenAI and Anthropic APIs
- Financial Datasets MCP- Financial data integration and analysis
- N8n Workflow Generator MCP- Workflow automation platform
- Docker Orchestration MCP- Container lifecycle management
- Knowledge Memory MCP- Vector-based persistent storage
- GitHub Integration MCP- Repository management and automation
- Firecrawl Custom MCP- Web scraping and content extraction
- ScreenPilot MCP- Advanced UI element detection and automation
- SQLite MCP- Local database operations
- Memory MCP (Official)- Standard MCP memory implementation
- Filesystem MCP- Secure file operations with path sandboxing
- Sequential Thinking MCP- Structured problem-solving framework
- Playwright MCP- Browser automation and testing
- Fantasy Premier League MCP- Sports analytics and data

- Python 3.11+orNode.js 18+
- Claude DesktoporClaude Code CLI
- FastMCP frameworkfor custom server development

git clone https://github.com/GrimFandango42/Claude-MCP-tools.git cd Claude-MCP-tools

# Install FastMCP framework pip install fastmcp python-json-logger # Install development tools (optional) pip install black prettier pip-audit safety

# Copy configuration template cp claude_desktop_config_template.json $CLAUDE_CONFIG_PATH # Edit configuration with your API keys and paths # Location: %APPDATA%\Claude\claude_desktop_config.json (Windows)

"Search memory for server status and show me what MCP servers are available"

All servers integrate with the Memory MCP system for persistent context:

- Entity Storage: Projects, servers, processes, and knowledge
- Relation Mapping: Connected information discovery
- Cross-Session Continuity: Context preservation across conversations
- Searchable Knowledge: Single-keyword search with comprehensive results

AgenticSeek MCP provides intelligent AI provider selection:

- Local Processing: Free, private DeepSeek AI (privacy priority)
- Cloud APIs: OpenAI GPT-3.5/4 (speed priority) and Google Gemini (cost priority)
- Smart Routing: Automatic provider selection based on task characteristics
- Cost Optimization: Transparent cost estimation and optimization

Comprehensive tools for AI-assisted development:

- Code Analysis: AST parsing, complexity metrics, symbol resolution
- Quality Assurance: Automated formatting, linting, and security scanning
- Task Delegation: Claude Code Integration for complex coding tasks
- Project Intelligence: Automated codebase analysis and recommendations

from fastmcp import FastMCP mcp = FastMCP("server-name") @mcp.tool() async def process_data(input: str, options: dict = None) -> dict: """Process data with specified options.""" # Server implementation return {"status": "success", "result": processed_data} if __name__ == "__main__": mcp.run(transport="stdio")

{ "mcpServers": { "server-name": { "command": "python", "args": ["path/to/server.py"], "cwd": "working/directory", "env": { "API_KEY": "your-api-key", "LOG_LEVEL": "INFO" }, "keepAlive": true, "stderrToConsole": true } } }

- Start sessionsby searching memory for relevant context
- Document discoveriesin memory during work
- Create relationsbetween connected components
- End sessionsby updating memory with outcomes

# Search for project context mcp__memory__search_nodes("server") mcp__memory__search_nodes("consolidation") mcp__memory__search_nodes("FastMCP") # Search by entity type mcp__memory__search_nodes("configuration") mcp__memory__search_nodes("documentation")

- FastMCP Framework: Use for all new MCP servers
- Async Patterns: Proper async/await implementation
- Error Handling: Comprehensive error management and logging
- Memory Integration: Document discoveries in Memory MCP
- Testing: Independent testing before Claude Desktop integration

- Logging: Structured logging to stderr only (stdout reserved for JSON-RPC)
- Type Hints: Full type annotation for better tooling
- Documentation: Rich docstrings for tools and functions
- Security: Input validation and secure API key management

{ "stderrToConsole": true, "env": { "MCP_LOG_LEVEL": "DEBUG", "CLAUDE_CODE_MOCK": "true" } }

June 2, 2025 - Comprehensive MCP Testing & Validation Complete ✅

- Full System Testing: 94/100 overall score with all 7 core tool categories working perfectly
- Live Tool Validation: Memory (27 entities), AgenticSeek (4 AI providers), Firecrawl, GitHub, SQLite operational
- Integration Workflows: 3/3 complex scenarios tested and verified functional
- Performance Excellence: Sub-second response times across all MCP tools
- Testing Framework: Automated testing suite created for continuous validation
- Comprehensive Memory Structure: 27 entities, 31 relations for project knowledge
- Claude Code Integration Testing: 100% success rate (2/2 tasks completed)
- Enhanced Documentation: Professional system instructions and workflow templates
- Server Consolidation: Optimized from 25 to 21 servers with improved efficiency

May 31, 2025 - AI Routing System Complete ✅

- AgenticSeek MCP Server: AsyncIO bug resolution with 100% operational success
- Multi-Provider Integration: Local DeepSeek, OpenAI GPT-3.5/4, Google Gemini
- Smart Cost Optimization: Intelligent provider selection based on task characteristics

Claude-MCP-tools/ ├── servers/ # MCP server implementations │ ├── agenticseek-mcp/ # Multi-provider AI routing │ ├── claude-code-integration-mcp/ # Claude Code CLI bridge │ ├── windows-computer-use/ # Desktop automation │ └── ... # Additional servers ├── docs/ # Comprehensive documentation ├── scripts/ # Deployment and testing scripts ├── archive/ # Historical and deprecated files ├── CLAUDE.md # Development guidelines ├── PROJECT_STATUS_UPDATED.md # Current project status └── CONFIG_UPDATE_COMPLETE.md # Configuration management guide

We welcome contributions! Please see our development guidelines:
- Follow FastMCP patternsfor server development
- Use Memory MCPto document discoveries and solutions
- Test independentlybefore Claude Desktop integration
- Update documentationwith new capabilities

- Documentation: Comprehensive guides in/docs/directory
- Development Guidelines: SeeCLAUDE.mdfor standards and patterns
- Project Status: Current status inPROJECT_STATUS_UPDATED.md
- Memory Workflow:MEMORY_WORKFLOW_GUIDE.mdfor context management
- Session Templates:SESSION_CONTEXT_TEMPLATE.mdfor structured work

Built on the Model Context Protocol specification by Anthropic. Leverages FastMCP framework for efficient server implementation and the official Memory MCP server for persistent context storage.

Status: Production Ready |Last Updated: June 2, 2025 |Servers: 21 Operational

This is a web browser that enables your coding agent, such as Claude Code, to visit websites on your behalf and assist you in identifying bugs or creating UI test cases.

Rewrites coding prompts for optimal results with AI IDEs like Cursor AI, powered by Claude by Anthropic.

An MCP server to help AI assistants to answer questions and generate AccelByte Extend SDK code more effectively .

Local stdio MCP server that lets AI coding agents read and maintain structured architecture, rules, and decisions directly from your repository.

Official Context7 MCP server that brings up-to-date, version-specific library documentation and code examples into AI coding prompts.

Remote, no-auth MCP server providing AI-powered codebase context and answers

Instead of direct calling MCP tools, mcpcode server transforms MCP tool calls into TypeScript programs, enabling smarter, lower-latency orchestration by LLMs.

next-devtools-mcp is a MCP server that provides Next.js development tools and utilities for AI coding assistants like Claude and Cursor.

Help agents automatically write and test stories for your UI components