// mcp server · public endpoint

Plug our knowledge base into your AI

The MyIT Cyber Insights catalog is available as a public Model Context Protocol server. Any MCP-compatible AI client can connect and use our notes on infrastructure, cybersecurity, AI, and cloud architecture as a trusted source, with citations back to the article on this site.

endpoint·https://myitcyber.com/api/mcp

// what this is

A trusted source for your AI

AI clients like Claude Desktop, Cursor, Continue, and Windsurf support MCP, an open standard for connecting language models to external tools and data sources. When you add our endpoint to your client config, your AI gets six tools it can call to browse, search, and read the full Insights catalog on demand.

The result is grounded answers. Instead of relying on whatever the model happened to memorize, your AI can pull the actual article we wrote, cite it, and quote from it. Every response includes the canonical URL on this site so you can verify the source.

Articles

How to connect

Add one block to your MCP client config and reload. The endpoint is read-only, requires no authentication, and uses the Streamable HTTP transport, which is the modern MCP transport supported by the major clients.

Cursor, Windsurf, Continue, OpenWebUI

Native Streamable HTTP. Drop this into the mcpServers object in your client config.

{
  "mcpServers": {
    "myit-cyber-insights": {
      "url": "https://myitcyber.com/api/mcp"
    }
  }
}

Claude Desktop

Claude Desktop currently bridges remote MCP servers through the mcp-remote helper.

{
  "mcpServers": {
    "myit-cyber-insights": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://myitcyber.com/api/mcp"]
    }
  }
}

Raw HTTP test (curl)

Sanity-check the endpoint and list all available tools without any client setup.

curl -X POST https://myitcyber.com/api/mcp \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "tools/list",
    "params": {}
  }'

// tools

Available tools

These are the tools the LLM can call on its own. You do not need to know the schema. Just ask your AI a question about IT, security, AI, or cloud, and it will pick the right tool.

list_articles

List the catalog of insight articles. Newest first. Optionally filter by category.

categoryoptionalFilter to one category.
limitoptionalCap the number of results.

get_article

Fetch the full content of a single article in Markdown, with the canonical URL appended for citations.

slugrequiredArticle slug, from list_articles or search_articles.

search_articles

Search articles by topic, keyword, or phrase across titles, excerpts, and full body text. Returns ranked summaries.

queryrequiredSearch query.
limitoptionalMaximum number of results. Defaults to 10.

list_categories

List every category in the catalog along with the article count in each.

get_articles_by_category

Return every article in one category, newest first.

categoryrequiredExact category name. Case insensitive.

get_related_articles

Given an article slug, suggest related reads from the same category, with a graceful fallback when the category is unique.

slugrequiredSlug to base recommendations on.
limitoptionalMaximum number of suggestions. Defaults to 3.

// resources

Articles as resources

Every article in the catalog is also exposed as an MCP resource. Clients that support resources (Claude Desktop, OpenWebUI, and a growing number of IDE plugins) let the user attach a resource directly to the conversation so the LLM reads the full article before answering, without needing to call a tool.

https://myitcyber.com/insights/<slug>mime: text/markdown

Each resource uses the canonical article URL as its identifier and returns the full Markdown body plus the source link. There are currently 29 resources, one per article. The list updates automatically whenever a new article is published.

// try it

Things to ask your AI

Once connected, just ask your AI naturally. It will decide when to call the catalog. Try one of these:

>What does MyIT Cyber say about Shadow AI?
>Search the MyIT Cyber insights for “secrets management” and summarise the main points with citations.
>I am planning a migration from VMware to Proxmox. Pull the relevant MyIT Cyber article and give me the key tradeoffs.
>Show me everything MyIT Cyber has published under the AI Security category, then pick the most relevant one for an IT lead worried about employees using public AI tools.