MikeCreighton.com Content MCP Server

MikeCreighton.com Content MCP Server

The MikeCreighton.com Content MCP server solves a common pain point for AI‑powered assistants: staying current with a dynamic, content‑rich website. By exposing every page of Mike Creighton’s consulting site as a separate, queryable resource, the server lets Claude and other MCP clients ingest up‑to‑date marketing material without manual copy‑and‑paste or frequent API calls. Developers can now treat the entire website as a structured knowledge base that updates automatically whenever the site changes, ensuring conversations reflect the latest services, blog posts, and client success stories.

At its core, the server offers three intuitive tools. returns a catalog of all available pages, enabling developers to browse the knowledge graph on demand. fetches the full Markdown content of a chosen page, while provides keyword search across titles and descriptions. Together, these tools give an assistant the ability to pull precise information—such as a specific service description or recent blog insights—directly into a dialogue, streamlining content discovery and reducing hallucination risk.

Key features include runtime resource generation: each page is turned into a named Resource, complete with Markdown formatting and metadata (title, description). This design allows developers to reference exact pages in prompts or use them as building blocks for composite queries. The server’s lightweight Python implementation, managed via , ensures zero‑setup virtual environments and fast deployment on any machine. The included utility automates the entire pipeline: it crawls the live site, converts HTML to Markdown with MarkItDown, and produces a JSON map that reflects the site’s hierarchy. This keeps the MCP resources fresh with every run.

Real‑world use cases are plentiful for agencies, marketers, and content strategists. An AI assistant can quickly answer “What services does Mike offer?” or generate a draft RFP response by pulling relevant pages and synthesizing their content. Marketing teams can prototype chatbots that surface the latest blog posts or case studies, while business development reps can use the server to prepare tailored outreach based on current website messaging. Because each page is a discrete resource, developers can also build fine‑grained access controls or incremental updates without re‑indexing the entire site.

Integration into existing AI workflows is straightforward. Once the MCP server is running, any Claude Desktop client can be configured to connect via the entry. From there, prompts that invoke the provided tools automatically fetch real content, allowing developers to focus on higher‑level logic—such as summarization, tone adjustment, or cross‑referencing—rather than on data ingestion. The server’s clear separation of tools and resources makes it a plug‑and‑play component in larger, multi‑source knowledge systems.

In summary, the MikeCreighton.com Content MCP server turns a static website into a dynamic, queryable knowledge base. Its simple toolset, automatic content extraction, and seamless integration with Claude make it an invaluable asset for developers looking to embed up‑to‑date, high‑quality content into AI assistants without manual maintenance.