Turn your Strapi content into AI tools with MCP, and add a TanStack AI chat that runs on your machine

Strapi 5.47+ includes an MCP server that can expose content-management actions to AI clients such as Claude and Cursor when enabled and authenticated with admin tokens, whose permissions determine which tools and content are visible. The tutorial demonstrates creating a Strapi project with seeded content, connecting an MCP client, and developing a reusable healthcheck service that verifies application and database status through three interfaces: a REST endpoint, a custom MCP tool, and an in-admin AI chat tool. It distinguishes services, controllers, routes, MCP tools, API tokens, admin tokens, public/authenticated roles, and administrator RBAC, emphasizing that each interface requires separate permissions. It also explains how to install the TanStack AI plugin and Ollama to run a local tool-capable model without sending content or API keys off the machine, then register app services as chat tools through configuration and metadata. For reusable tools, the tutorial recommends packaging them as plugins, which centralizes tool definitions, permissions, MCP registration, and automatic chat discovery, while troubleshooting guidance focuses on permission misconfiguration, authentication-token misuse, tool registration, and stale client tool lists.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.