Skip to main content
Glama
nobitalqs

Modular RAG MCP Server

by nobitalqs
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Hybrid

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault

No arguments

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": false
}
experimental
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
query_knowledge_hubB

Search the knowledge base for relevant documents.

This tool uses hybrid search (semantic + keyword) to find the most relevant documents matching your query. Results include source citations for reference.

Parameters:

  • query: Your search question or keywords

  • top_k: Maximum number of results (default: 5)

  • collection: Limit search to a specific document collection

list_collectionsA

List all available document collections in the knowledge base.

Returns information about each collection including:

  • Collection name

  • Document count (if include_stats=true)

  • Collection metadata

Use this tool to discover available collections before querying.

get_document_summaryA

Get summary and metadata for a specific document.

Returns structured information about a document including:

  • Title (extracted or inferred from content)

  • Summary (first chunk preview or metadata summary)

  • Tags (document-level tags/categories)

  • Source path

  • Chunk count

Use this tool after list_collections to get details about specific documents.

delete_documentA

Delete a document and its associated data from the RAG knowledge base.

First call without confirm_delete_data (or with confirm_delete_data=false) to preview what will be deleted (chunk count, image count). Then call again with confirm_delete_data=true to execute the deletion.

This removes all associated data: vector embeddings, BM25 index entries, extracted images, and the ingestion history record.

ingest_documentA

Ingest a document file into the knowledge hub.

Supports PDF, Markdown (.md/.markdown), and source code (.py, .c, .cpp, etc.). The document is parsed, chunked, embedded, and stored for later retrieval.

Parameters:

  • file_path: Absolute or relative path to the file to ingest

  • collection: Target collection name (default: "default")

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Deploy Server
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

TDQS

A4/5.0

Scored across 5 tools

Disambiguation5/5

Each tool serves a unique purpose: ingestion, deletion, listing collections, querying, and document summary retrieval. No functional overlap exists between these tools.

Naming Consistency5/5

All tool names follow the verb_noun pattern consistently (e.g., delete_document, list_collections). No mixing of camelCase or other conventions.

Tool Count5/5

With 5 tools, the server covers the essential operations for a RAG knowledge base: add, remove, list, detail, and search. The count is well-scoped without bloat.

Completeness4/5

Covers core CRUD and query operations. Missing an explicit tool to list all documents within a collection (only collections are listed), and no update tool (though delete+re-ingest is a workaround). Minor gaps but overall sufficient.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources