Monta Knowledgebase MCP Server

Welcome to Monta's Knowledgebase MCP (Model Context Protocol) server. Access comprehensive documentation, help articles, and technical resources through AI-powered search.

Contact us here for issues

🚀 What is MCP?

Model Context Protocol (MCP) is an open protocol that standardizes how applications provide tools, resources, and contextual data to AI systems. This server exposes Monta's knowledge base as searchable MCP tools for AI assistants and automation platforms.

🏗️ What You Can Do

Search and retrieve Monta documentation using:

🔍 Knowledge Base Content

Our knowledge base includes comprehensive content from multiple sources:

Source Description
Help center User guides and troubleshooting
Support answers +250K answered tickets
Product Documentation Technical documentation
Product Updates Release notes and changelogs
Industry documents OCPP, OCPI & more

Data Freshness: The knowledge base is automatically synced with fresh data on a weekly basis to ensure you have access to the latest documentation, support answers, and product updates.

Check the query logic →

📖 Intelligent Glossary System

The knowledge base includes an intelligent glossary system that automatically normalizes alternative terminology to canonical terms during search, improving search accuracy and consistency.

How it works:

Example normalizations:

"charger" → "charge point"
"admin" → "Manager"
"AC Charger" → "AC charge point"
"ev" → "Electric Vehicle"
"CPO" → "Charge Point Operator"

Try it yourself: Visit the Interactive Query Tester and search for "charger" - you'll see how it's normalized to "charge point" with a visual indicator showing the transformation.

Advanced usage: Set skipGlossaryNormalize=true to bypass glossary normalization if you need to search for exact terms.

🔧 Available Tools

The server provides MCP tools for AI-powered documentation access:

query - Search Documentation

Search Monta documentation with multiple modes and filters. English queries only.

Parameters:

Response Format: Each result includes:

⚠️ Important: Content is automatically truncated to 3000 characters for efficiency. If truncated: true, use the fetch tool to retrieve the complete document.

fetch - Get Document by ID

Retrieve the complete, untruncated document by its ID. Use this when search results have truncated: true.

Parameters:

Response Format: Returns the full document with all fields and complete content (no truncation).

Browse Interactive Tools →

📚 Available Resources

Access documentation resources directly:

Resource Description
monta://resources/health Current server status and statistics
monta://resources/query/{query} Search documentation by query
monta://resources/fetch/{document_id} Retrieve a specific document by ID

Browse Interactive Resources →

💡 Available Prompts

Pre-built workflow templates for common tasks:

Browse Interactive Prompts →

🔐 Quick Setup

Claude Desktop (Remote)

Choose one of the transport methods below:

Option 1: Streamable HTTP (Recommended)

{
  "mcpServers": {
    "monta-knowledgebase": {
      "url": "https://knowledgebase.monta.app/mcp",
      "transport": {
        "type": "streamable-http"
      }
    }
  }
}

Option 2: SSE (Server-Sent Events)

{
  "mcpServers": {
    "monta-knowledgebase": {
      "url": "https://knowledgebase.monta.app/sse",
      "transport": {
        "type": "sse"
      }
    }
  }
}

n8n (MCP Client Tool Node)

Option 1: Streamable HTTP (Recommended)

Option 2: SSE (Server-Sent Events)

REST API

Try the Interactive Query Tester →

Query documentation (GET):

# Default mode (hybrid search)
curl "https://knowledgebase.monta.app/api/query?query=charging%20stations&maxResults=5"

# BM25 mode (keyword search only)
curl "https://knowledgebase.monta.app/api/query?query=charging%20stations&maxResults=5&mode=bm25"

# Vector mode (semantic search only)
curl "https://knowledgebase.monta.app/api/query?query=charging%20stations&maxResults=5&mode=vector"

# Filter by source
curl "https://knowledgebase.monta.app/api/query?query=OCPP&source=helpjuice&maxResults=10"

# Filter to factual sources only (official documentation)
curl "https://knowledgebase.monta.app/api/query?query=charging%20stations&factual=true&maxResults=10"

# Bypass language check (advanced)
curl "https://knowledgebase.monta.app/api/query?query=ladestation&skipLanguageCheck=true&maxResults=5"

Query documentation (POST):

curl -X POST https://knowledgebase.monta.app/api/query \
  -H "Content-Type: application/json" \
  -d '{
    "query": "charging stations",
    "maxResults": 5,
    "mode": "default",
    "source": "helpjuice"
  }'

Available query parameters:

Fetch document by ID:

# Fetch a single document without embeddings (lightweight)
curl "https://knowledgebase.monta.app/api/fetch/9a4af45c43e3fab0f763669534fa39ba"

📖 Documentation

For detailed usage instructions and integration guides, see: