Showroom Docs MCP Server

Quarkus MCP Server that indexes Red Hat product documentation and the "IA Development From Zero To Hero" workshop for OpenShift Lightspeed.

Project maintained by maximilianoPizarro Hosted on GitHub Pages — Theme by mattgraham

Architecture

English | Espanol

Architecture: OpenShift Lightspeed + MCP

Runtime: Quarkus 3.27.3 with Java 21
Protocol: MCP via Streamable HTTP (/mcp)
Content: 46 embedded markdown documents (~3MB), auto-sanitized on load
Port: 8080
Sanitization: PDF artifacts (page numbers, legal notices, TOC) removed automatically
Smart truncation: Content cut at paragraph boundaries, no visible truncation messages

The user asks a question in the OLS console
OLS API Server receives the query and sends it to the LLM
The LLM invokes showroom-docs-mcp tools for documentation queries
The MCP server searches 46 sanitized documents and returns relevant sections
Content is auto-sanitized (PDF artifacts removed) and smart-truncated at paragraph boundaries
The LLM generates a contextualized response using the retrieved documentation

Tool	Description	Input
`searchDocs`	Keyword search across all documentation	`query: string`
`listDocSections`	List all available document sections	(no parameters)
`getDocSection`	Get full content of a specific section (supports fuzzy matching)	`fileName: string`
`getDocSummary`	Get knowledge base overview and example questions	(no parameters)

The LLM decides which tools to call based on the question:

Question Type	Expected Tool
“How do I install Service Mesh?”	`searchDocs` → searches documentation
“List all available documentation”	`listDocSections` → returns index
“Show me the Developer Hub docs”	`getDocSection` → returns full doc
“What products are indexed?”	`getDocSummary` → returns overview
“How do I install Ansible?”	`searchDocs` → searches developer docs
“Quick-start for Quarkus?”	`searchDocs` → returns CLI commands and code

MCP Tool Selection Flow

Knowledge Base: 46 Indexed Documents