MCP
MCP Sections
- Web Model Context Protocol
- Declarative API
- Imperative API
- Agentic Web
- Natural Language SQL
- Autonomous Research Agents
- Obsidian MCP
- Home Assistant MCP
- Sandboxed File Management
- Developer Knowledge API
- Model Context Protocol (MCP)
- Google Cloud
- Firebase
- Android
- Agentic Tools
- Documentation API
- Model Context Protocol (MCP)
- MCP Connectors
- Agentic Integration
- Tool Discovery
- JSON-RPC
- Supabase MCP
- Chrome DevTools MCP
- GitHub MCP Registry
- Model Context Protocol (MCP)
- MCP Server
- mcp-publisher CLI
- Enterprise Governance
- Developer Tools
- Agentic AI
- local LLM
- MCP server
- on-premises AI
- data sovereignty
- agentic infrastructure
- Ollama
- Model Context Protocol
- client-server architecture
- agentic AI
- Host application
- MCP Client
- MCP Server
- dynamic context discovery
- context window optimization
- agentic tooling
- command-line interface
- MCP servers
Foundational knowledge asset for Google WebMCP. Defines the protocol's declarative and imperative integration paths, its role in replacing vision-based screen scraping with structured exchanges, and the performance and security benefits of structured agent-website communication. Turns Chrome into an AI agent surface where sites publish capabilities directly to agents instead of requiring fragile pixel-level interpretation.
Five specific use cases for Model Context Protocol (MCP) with local LLMs: natural language database querying (SQLite, PostgreSQL, MySQL), autonomous web research, Obsidian vault management, offline smart home control via Home Assistant, and natural language file system management. Each scenario shows how a local LLM gains capability through tool access without sacrificing privacy or incurring API costs.
The Google Developer Knowledge API and its official Model Context Protocol (MCP) server. These tools provide a canonical, machine-readable gateway to Google's public developer documentation for platforms like Firebase, Android, and Google Cloud. By connecting to the MCP server, AI assistants can retrieve fresh, accurate documentation as Markdown, enabling more reliable implementation guidance and troubleshooting.
How Model Context Protocol (MCP) connectors serve as the bridge between AI agents and external systems. Covers the architecture, common integration patterns (local vs. remote), and provides examples of connectors for databases (Supabase), developer tools (Chrome DevTools), and platforms (WordPress). Outlines best practices for security, discovery, and development using various SDKs.
The GitHub MCP Registry, a central hub for discovering, installing, publishing, and managing MCP servers. Covers the `mcp-publisher` CLI for publishing, automated workflows with GitHub Actions, and enterprise governance features like allow lists for secure tool usage in AI development. Provides a structured ecosystem for organizations to govern which MCP tools their developers and agents may use.
A comprehensive guide to integrating Model Context Protocol (MCP) servers with locally hosted Large Language Models (LLMs). Covers hardware and software requirements, deployment architecture, security considerations, and performance optimization strategies for creating secure, private, and scalable on-premises AI systems.
Defines the Model Context Protocol (MCP) as an open standard for agentic AI. Details the client-server architecture, explaining the distinct roles of the Host (user-facing application), the Client (communication handler), and the Server (external tool provider), which together enable AI models to securely and dynamically interact with external data and functions.
The MCP CLI is a lightweight command-line interface designed to solve the context window bloat problem in agentic AI systems. Enables dynamic, just-in-time discovery of MCP server tools, drastically reducing token consumption compared to static schema loading. Covers installation, configuration, and integration patterns for AI agents that need to invoke many tools without preloading all schemas.