azure-ai-search-foundry-iq-live-knowledge-sources
Active·★ 17·MIT·Updated 2026-10-13
This accelerator helps platform teams connect various knowledge sources, such as Fabric Ontology or remote MCP tools, to an Azure AI Search Foundry IQ Knowledge Base. It enables natural language querying to receive grounded results with verifiable source evidence, integrating seamlessly with both REST and MCP clients.
#Azure AI Search#Foundry IQ#Knowledge Base#RAG#Fabric Ontology#MCP#Agentic Retrieval#LLM Integration
01
Features
01Integrates diverse knowledge sources, including Fabric Ontology and remote HTTPS MCP servers, into a unified Knowledge Base.
02Supports natural language queries against the Knowledge Base, providing grounded results with activity, references, and source data.
03Offers a comprehensive CLI (LiveKS) for validation, deployment, verification, invocation, and cleanup of resources.
04Exposes a native `knowledge_base_retrieve` endpoint compatible with MCP clients for seamless integration into agentic workflows.
05Includes an offline 'try' mode to inspect response and evidence contracts without requiring cloud resources.
02
Compatibility
Python
Runtime
Verified via docs
Node.js
Runtime
Verified via docs
Azure AI Search
Platform
Verified via docs
Microsoft Fabric
Platform
Verified via docs
MCP
Protocol
Verified via docs
Bash & PowerShell
CLI Environment
Verified via docs
03
Quick start
1
$ git clone --depth 1 https://github.com/microsoft/azure-ai-search-foundry-iq-live-knowledge-sources.git
2
$ cd azure-ai-search-foundry-iq-live-knowledge-sources
3
$ ./liveks bootstrap
4
$ ./liveks init --profile byo-fabric --env liveks-byo
04
Use cases
↳Evaluating Foundry IQ Live Knowledge Sources for managed organizations and conducting field demonstrations.
↳Connecting existing Microsoft Fabric workspaces and ontologies to enhance knowledge base grounding.
↳Validating live MCP Server Knowledge Sources to integrate external tools into retrieval processes.
↳Rehearsing greenfield deployments of Fabric and Azure AI Search resources for agentic retrieval solutions.
05
Alternatives
seekstone★ 34
The Obsidian MCP server that needs no plugin, no running Obsidian app — and doesn't blow your context window. Filesystem-direct, single-digit-ms search, ~2 KB payloads, 17 tools.
answerLoops★ 22
Agent-native AI support that turns your docs into answers for your community. Built for your agents with skills and MCP.
rosetta★ 50
MCP Server with RouterOS docs + commands + products + changelogs, using SQLite-as-RAG, sourced from MikroTik
company-docs-mcp★ 49
AI-powered company knowledge MCP. Unified place for internal policies, values, documentation, and governance. Agents can search, cite, and answer questions using real company docs.
obsidian-mcp-server★ 19
MCP server for Obsidian vault management - enables Claude and other AI assistants to read, write, search, and organize your notes
Comments
Log in to leave a comment
No comments yet. Be the first!