I did a celebratory air punch when I read through the JWT/IAM additions. Huge addition to the product.

I'm so glad in my recent episode of Knowledge Graph Insights with Larry Swanson, that I finally got to talk about something that's been overlooked - The timeline of the explosion of the term context graph. Yes, it started with the Foundation Capital article, but what happened next? I wouldn't have even known about that article if it weren't for @KirkMarple and the response his rebuttal article got. But what was really going on? Check out my conversation with Larry for my take on why the term context graph exploded in over the holidays of 2025. Audio, video, and full transcript: knowledgegraphinsights.com/d…

How ontologies like W3C PROV-O are the foundation of explainable AI.

TrustGraph - Graph-native AI context platform (@TrustGraphAI) Cossmology Profile: dub.sh/yDcLAoc Key People: Daniel Davis (@TrustSpooky), Mark Adams (@Cybermaggedon) #KnowledgeGraph #OpenSource #OSS #COSS

What are ontologies and how are they relevant to context graphs?

Another milestone! 🎉 🌟 Not long ago, we celebrated 1,000 🌟 — and the momentum since then has been incredible. After spening nearly an entire week trending on GitHub, we celebrate 2,000 🌟! This isn't just a vanity metric. Every star represents a developer, researcher, or builder who saw TrustGraph and thought, "yes, this is the context layer AI agents have been missing." That means everything to us. We're building the infrastructure that makes AI agents actually trustworthy — grounded in structured context, not hallucinated guesses. The fact that the community is rallying around that vision is fuel for everything we're doing next. Thank you to every contributor, early adopter, and supporter who helped get us here. The best is still ahead. 🚀 ⭐github.com/trustgraph-ai/tru…

Most AI stacks hide a dirty secret: swap your LLM provider, and everything breaks. TrustGraph was built differently. TrustGraph ships with a full, modular LLM inferencing stack — so you're never locked into a single model or provider. Whether you're running fully air-gapped on-prem or mixing cloud APIs, TrustGraph has you covered out of the box. The inferencing stack supports: 🖥️ Local / Self-Hosted — vLLM, TGI, Ollama, LM Studio, and llama.cpp for full on-premises deployments ☁️ Cloud Providers — OpenAI, Anthropic, Google Gemini, AWS Bedrock, Azure OpenAI, Cohere, and more 🔧 Fully Configurable — swap providers without rewiring your entire pipeline This isn't just about flexibility. It's about control — over cost, latency, data privacy, and compliance. In regulated industries especially, the ability to keep data on your own infrastructure while still leveraging the best available models is a game-changer. TrustGraph's context graph operating system handles the heavy lifting — knowledge extraction, entity resolution, semantic retrieval — and hands off to whatever LLM you trust most. Your stack. Your rules. 👉 Check out the leading context graph operating system: github.com/trustgraph-ai/tru…

The biggest problem with RAG in production: you don't know why your AI said that. TrustGraph 2 fixes that with full-stack explainability. Here's how it works At ingest time, TrustGraph emits PROV-O triples tracing every fact back to its source — document → page → chunk → graph edge. Full lineage, built in. At query time, every GraphRAG, DocumentRAG, and Agent call records a complete reasoning trace: question formation, grounding, exploration, focus, synthesis. All stored in a dedicated named graph you can query or export. New CLI tools let you list and inspect every trace: tg-list-explain-traces, tg-show-explain-trace, tg-show-extraction-provenance. Or use the new Workbench Explainability Panel. TrustGraph 2 also ships RDF-star support, a redesigned entity-centric graph schema, pluggable Tool Services for agents, batch embeddings, and streaming triple queries. It's open source and runs anywhere. 🔗github.com/trustgraph-ai/tru…