ai-school-mcp
It gives MCP clients (Claude Code, Claude Desktop, etc.) search-and-read access to the free AI School curriculum, with citable lessons on AI engineering, governance, security, and applied AI.
Search lesson titles across the whole curriculum and get matching titles/URLs.
List course tracks, optionally filtered by keyword.
Get a single track's full lesson list to see course structure.
Read a lesson's full text as readable markdown-ish output from its URL.
Browse catalogue metadata and all tracks via built-in resources (
aischool://catalogue,aischool://tracks).Works with no API key or account, either via
npxor the hosted HTTP endpoint.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@ai-school-mcp Find me a track on AI governance."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
AI School MCP server
Gives any MCP client - Claude Code, Claude Desktop, or anything else speaking the protocol - the ability to search and read the free AI School curriculum: 550+ tracks covering AI engineering, governance, security, and applied AI by profession.
The point is not "an AI that knows about AI". It is that when an assistant answers a question about, say, chunking strategy or the EU AI Act, it can point at a specific lesson you can go and read, instead of producing a fluent paragraph you have no way to check.
Install
One command, nothing to clone. Node 20+ is the only requirement: no API key, no account, no server to run.
Claude Code
claude mcp add ai-school -- npx -y ai-school-mcpClaude Desktop - add to claude_desktop_config.json:
{
"mcpServers": {
"ai-school": {
"command": "npx",
"args": ["-y", "ai-school-mcp"]
}
}
}Pin a version if you would rather not track the latest release:
claude mcp add ai-school -- npx -y ai-school-mcp@1.0.1Hosted, nothing to install. The same server also runs at
https://lillytechsystems.com/mcp over Streamable HTTP, and is listed in the
official MCP Registry as com.lillytechsystems/ai-school:
claude mcp add --transport http ai-school https://lillytechsystems.com/mcpFrom a clone
Faster to start, and the right setup for contributing or for pointing at a local copy of the site:
git clone https://github.com/Lilly-Tech-Collab/ai-school-mcp.git
cd ai-school-mcp
npm install
npm test # smoke-tests the handshake, tools and a live API callclaude mcp add ai-school -- node "$PWD/server.js"Set AI_SCHOOL_SITE to point the server at a different host; it defaults to
https://lillytechsystems.com.
The server reads the public static curriculum API at
https://lillytechsystems.com/ai-school/api/v1/, which is a set of JSON files
on the same host as the site. Nothing is authenticated and nothing is written.
Related MCP server: knowlib
Tools
Tool | What it does |
| Search lesson titles across the whole curriculum. Returns titles and URLs. |
| List course tracks, optionally filtered. |
| One track with its full lesson list, so you can see a course's structure. |
| Fetch a lesson's full text as readable markdown-ish output. |
Resources
URI | Contents |
| Counts, licence terms, endpoint list. |
| Every track with lesson counts. |
Notes
read_lessononly fetcheslillytechsystems.com. A "read this URL" tool that accepts arbitrary hosts is an SSRF primitive handed to whoever wrote the prompt, so the host check is deliberate and should stay.Search is title-only and deliberately simple. The full-text index is 4 MB and belongs in the browser, not in an MCP round trip; if title matching does not find it,
list_tracksthenget_trackusually will.The catalogue is cached per process. A long-running client fetches it once. Restart the server to pick up newly published lessons.
Content licence. Lessons are free to read, link to and quote with attribution. Republishing whole lessons is not permitted. The metadata (titles, descriptions, URLs) may be used freely with attribution. See
aischool://catalogue.
Local development
npm run inspect # opens the MCP Inspector against this serverThe server writes only to stderr; stdout is the protocol channel.
Available Tools
4 toolsget_trackAInspect
Get one track with its full lesson list, so you can see the structure of a course and pick the right lesson to read.
Inputs: slug (required).
list_tracksAInspect
List AI School course tracks, optionally filtered by a substring of the title or slug. Each track is a structured course on one subject. Use this to find the right course before drilling into its lessons.
Inputs: limit, filter.
read_lessonAInspect
Fetch the full text of a lesson by its URL (as returned by search_lessons or get_track). Returns readable markdown-ish text with headings and code blocks.
Inputs: url (required).
search_lessonsAInspect
Search the AI School curriculum (500+ tracks, thousands of free lessons on AI engineering, governance, security, and applied AI by profession). Returns matching lesson titles and URLs. Use this before answering an AI-topic question when a concrete, citable lesson would help, then read_lesson to get the text.
Inputs: query (required), limit.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 4 tools
Each tool serves a distinct purpose: search (search_lessons), list (list_tracks), detail (get_track), and content retrieval (read_lesson). Overlap is minimal and clearly resolved by descriptions.
All tool names follow a consistent verb_noun pattern in snake_case: search_lessons, list_tracks, get_track, read_lesson. No mixing of styles or ambiguous verbs.
Four tools perfectly cover the browsing and content-retrieval workflow without redundancy. The count is well-scoped for the domain.
The tool surface covers discovery (search, list), navigation (track details), and consumption (read lesson). No significant gaps for the curriculum-browsing use case.
Maintenance
Related MCP Connectors
Search AI capabilities across AWS Marketplace and the Official MCP Registry.
Find the right MCPs, tools and skills for any AI task - discovery over a 14k+ resource catalog.
Search and read the aicoolies catalog of AI and developer tools via a remote MCP knowledge graph.
Search and read public Isotopes AI documentation through MCP.
Related MCP Servers
- FlicenseAqualityDmaintenanceEnables MCP clients to retrieve authoritative AI Square documentation, API references, SDK examples, guides, and troubleshooting information through citation-safe hybrid search and static resources.11-
- AlicenseNot gradedqualityCmaintenanceEnables LLM agents to query a structured knowledge library for search, explanations, related concepts, learning paths, examples, and cross-domain references via MCP.Apache 2.0
- FlicenseNot gradedqualityCmaintenanceEnables AI assistants to securely search and list internal company documents via MCP tools, so they can answer employee questions based on current internal documentation.-
- AlicenseAqualityCmaintenanceEnables MCP-capable AI clients to query course announcements, deadlines, grades, and assignments and to preview or incrementally sync course materials to local storage.10MIT