Kingsley Uyi Idehen

Semantic Web and LLM + MCP symbiosis

LLMs + the Model Context Protocol (MCP) are the Yang to the Semantic Web Project's Yin.

We now have a solution to the final hurdle—visualization.

Years of Linked Data work now come alive. I explain this, with demonstrations, in a new newsletter post.

www.linkedin.com/pulse/semant...

#MCP

MCP is like Unix Pipes Explainer

Page metadata snippet

The Model Context Protocol (MCP) brings Unix-style loose coupling to AI Agents.

I built a visual explainer (JSON-LD metadata included) using Google Gemini’s new Canvas feature—with an embedded Knowledge Graph to illustrate.

Check it out […]

[Original post on mastodon.social]

Sample message exchange, in chat form.

What’s going on here?

Prompts are being processed through queries executed against a Knowledge Graph (KG)—all thanks to loose coupling. The target KG could reside locally, live on the Web, or be part of the vast Linked Open Data (LOD) cloud.

#A2A #MCP #VirtuosoRDBMS #AI #Agentic #GenAI

A2A and MCP interaction

Architecture? As depicted in the attached image.

BTW -- @Google Gemini can generate pretty accurate diagrams using @drawio XML!

#A2A #MCP #OPAL #VirtuosoRDBMS #AI #Agentic

A simple GUI-based screencast demo of the @openlink AI Layer (OPAL), showing what’s now possible via MCP + A2A—namely, composing and orchestrating agentic workflows through loosely coupled software.

How?

An Agent Host routes tasks to Agents via A2A. Agents […]

[Original post on mastodon.social]

GitHub - OpenLinkSoftware/mcp-jdbc-server: Java based Model Context Procotol (MCP) Server for JDBC Java based Model Context Procotol (MCP) Server for JDBC - OpenLinkSoftware/mcp-jdbc-server

Here's a link to the Claude Desktop artifact.

https://claude.ai/public/artifacts/19f792bc-867a-4ef7-910a-4696bdeec7d4

Here's a link to the MCP server for JDBC repo.

https://github.com/OpenLinkSoftware/mcp-jdbc-server/

Employee Dashboard example

An example of what’s now possible—thanks to our (@openlink) sophisticated MCP Server for JDBC, Oracle’s native JDBC Driver, and Anthropic's Claude Desktop as the MCP client:

A dashboard generated from Oracle’s HR database using the following natural language […]

[Original post on mastodon.social]

Customer and Employee performance dashboard.

Here’s another dashboard variant built using Claude Desktop as the MCP client and the virtual Northwind Knowledge Graph, derived from underlying Northwind tables. This edition adds both a Company and Employee dimension […]

[Original post on mastodon.social]

Sales Performance Dashboard

This week’s edition of my @linkedin newsletter shines a light on the impact of the Model Context Protocol (MCP) across multiple computing eras—leading right up to today.

We’ve arrived at a moment where loose coupling principles are cool (and super useful) […]

[Original post on mastodon.social]

The Model Context Protocol acts as a universal translator between LLMs and data sources, eliminating complex platform-specific requirements. Our new open source MCP server for ODBC (mcp-odbc-server) enables seamless integration of any ODBC-accessible […]

[Video] [Original post on mastodon.social]

Screenshot depicting Medium's Data Export service promise.

@medium,

I still don't have an email message or any sense of the location of my export archive. We are now going on 48 hrs since my initial request.

/cc @evhead @biz

Medium (@medium@me.dm) 370 Posts, 212 Following, 20.4K Followers · An open platform for writing.

@medium I've requested a data export over 24 hrs ago. I am still awaiting email notification for the download link.

How can we prepare for a future migration from Bluesky to Mastodon? Bridgy Fed — a service that enables you to connect together your website, fediverse account and Bluesky account — will help. But Bridgy Fed needs to be easier to use. So I recruited Claude’s new Sonnet 7 model to help me do […]

Test 2:
What is https://x.ai/blog/grok-3#this about? Provide your response in JSON-LD comprising the following:

1. Blog/Article Description
2. FAQ
3. Glossary
4. HowTo
5. Relevant Product/Service, Offers, Features descriptions

You should set @base to the article url, and incorporate relevant […]

Attached is a Grok-based screencast showcasing how note-taking is evolving, now that LLMs eliminate the tedium of crafting machine-computable notes in RDF using any of its associated notations (e.g., JSON-LD).

LLMs are the “Easy Button” for understanding and harnessing the powerful benefits envisioned by The Semantic Web Project.

If you closely examine the GenAI disruption in software, you’ll notice that all roads lead to better note-taking—progressively enriching knowledge repositories that […]

AI Agent interaction with Uniprot Protein Knowledge Graph

Here’s a DataTwingler AI Agent configuration in Markdown, designed for interacting with UniProt’s massive Knowledge Graph, which consists of 210 billion machine-computable statements of fact, meticulously curated by domain experts.

github.com/OpenLinkSoft...

Here’s a DataTwingler AI Agent configuration in Markdown, designed for interacting with UniProt’s massive Knowledge Graph, which consists of 210 billion machine-computable statements of fact, meticulously curated by domain experts […]

[Original post on mastodon.social]

Links:

* https://sparql.uniprot.org/.well-known/void -- Knowledge Graph Stats

* https://sparql.uniprot.org/sparql -- Query Service Endpoint

* https://github.com/sib-swiss/sparql-training/tree/master/uniprot -- SPARQL Querying Tutorial

* […]

[Original post on mastodon.social]

UniProt’s latest publicly accessible Knowledge Graph now contains 210,122,358,019 RDF triples—powered by our (@openlink) #VirtuosoRDBMS open-source edition, live on the Web and available to AI Agents, Applications and Services across life sciences, genomics, bioinformatics, molecular biology […]

Here’s a DataTwingler AI Agent configuration in Markdown, designed for interacting with UniProt’s massive Knowledge Graph, which consists of 210 billion machine-computable statements of fact, meticulously curated by domain experts […]

[Original post on mastodon.social]

Links:

* https://sparql.uniprot.org/.well-known/void -- Knowledge Graph Stats

* https://sparql.uniprot.org/sparql -- Query Service Endpoint

* https://github.com/sib-swiss/sparql-training/tree/master/uniprot -- SPARQL Querying Tutorial

* […]

[Original post on mastodon.social]

UniProt’s latest publicly accessible Knowledge Graph now contains 210,122,358,019 RDF triples—powered by our (@openlink) #VirtuosoRDBMS open-source edition, live on the Web and available to AI Agents, Applications and Services across life sciences, genomics, bioinformatics, molecular biology […]

Variant of the same demo, but this time courtesy of Private LLM for iOS and macOS.

#AI #GenAI #LLMs

The pendulum is swinging back toward the edge of the global connectivity mesh we call the #Internet—a familiar cycle, much like the shift from time-shared mainframes to the PC and Mac era.

Microsoft's Phi-4 LLM running on my MacBook Pro (M1), courtesy of LM […]

[Original post on mastodon.social]

Quick evaluation test screencast and a tabulated offer analysis for the new GPT-o3 LLM family from #OpenAI.

#GenAI #AI

The latest edition of my newsletter is out!

This week, I focus on a deceptively simple reasoning and inference challenge to compare the capabilities of AI+Compute inference services from #OpenAI, #Google, #Microsoft, #Groq, #Alibaba (Qwen), #DeepSeek, and […]

[Original post on mastodon.social]

Project link: https://app.heygen.com/share/d9f453ed341c4369b0238940130e6481

NotebookLM was impressive, but now we have the visual edition from HeyGen.Check out this fantastic video podcast on the OpenLink AI Layer (#OPAL)—an add-on t... OpenLink AI Layer Overview -- Video Podcast via HeyGen

NotebookLM was impressive, but now we have the visual edition from HeyGen.ai

Check out this fantastic video podcast on the OpenLink AI Layer (OPAL)—an add-on to our Virtuoso Data Spaces platform.

https://youtu.be/ceBgsOW2EQI

#AI #GenAI #HowTo #KnowledgeGraph #OPAL #Video #Podcast

Q: Do large language models (LLMs) possess reasoning and inference capabilities?

A: Yes-ish.

The issue is that “yes-ish” isn’t enough for scaling an AI + compute business model. LLMs will struggle to evolve toward AGI without integrating Knowledge Graphs via RAG or GraphRAG, where reasoning […]