# context-graph Published articles for context-graph. This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## V7 uses a Context Graph to give AI agents institutional memory DevFeed: [V7 uses a Context Graph to give AI agents institutional memory]() Original publisher: [Read original article]() Published: 2026-09-21T00:00:00Z Content type: article Language: en Sources: [OpenAI News]() Topics: [Artificial Intelligence](), [context](), [Graphs](), [Model Context Protocol](), [Repository](), [Low code]() Tags: [agents](), [ai](), [ai-agents](), [context](), [context-graph](), [graph](), [mcp](), [repositories](), [startup]() ### AI overview V7 uses GPT-5.6 and a Context Graph to organize information from scattered company files, connect entities and relationships, and support source-linked AI workflows. The platform is intended for complex enterprise tasks across finance, insurance, and real estate. ### Source excerpt Using GPT-5.6, V7 turns scattered company files into context agents can use to complete complex, source-linked work. ## From Agent Memory to Portable Skills DevFeed: [From Agent Memory to Portable Skills]() Original publisher: [Read original article]() Author: Jocelyn Hoppa Published: 2026-08-06T14:35:00Z Content type: article Language: en Sources: [Neo4j]() Topics: [Agent Skills](), [Neo4j](), [context](), [long-context](), [tracing](), [incident]() Tags: [agent](), [agent-memory](), [agentic-ai](), [context-graph](), [developer](), [genai](), [memory](), [neo4j](), [neo4j-agent-memory](), [ontology](), [procedural](), [service](), [skills](), [traces]() ### AI overview This article explains how Neo4j Agent Memory Service turns procedural knowledge from an organization's context graph into portable, provenance-grounded Agent Skill packages. It distinguishes persistent memory from actionable procedures that fresh agents can load and execute. ### Source excerpt Grounded skill distillation on a context graph in the Neo4j Agent Memory Service (NAMS) An agent accumulates hard-won procedural knowledge inside a single workspace and single context window: how to triage a P1 incident, which tool to call first, or what... Read more -> ## What is a context graph? DevFeed: [What is a context graph?]() Original publisher: [Read original article]() Author: Alyssa Di Pasqualucci Published: 2026-08-04T17:12:29Z Content type: tutorial Language: en Sources: [Neo4j]() Topics: [AI Agent](), [context](), [Graphs](), [sessions](), [audit trail](), [execution](), [systems]() Tags: [agent](), [agentic-ai](), [ai-agent](), [audit-trail](), [context](), [context-graph](), [execution](), [graph](), [sessions]() ### AI overview This guide explains context graphs as persistent memory systems for AI agents. It describes how they connect enterprise knowledge, conversation history, tool results, decisions, and reasoning so agents can maintain context across sessions, make more consistent decisions, and preserve audit trails. ### Source excerpt If you ask an AI agent the same question twice and get a different answer, that's a major issue. Without durable memory, context may not carry over between sessions. A customer support agent can approve a refund, lose the evidence... Read more -> ## One starter repo, three winners: Neo4j at Hack the Video Agent Context Graph DevFeed: [One starter repo, three winners: Neo4j at Hack the Video Agent Context Graph]() Original publisher: [Read original article]() Author: Jeremy Adams Published: 2026-08-01T04:08:00Z Content type: article Language: en Sources: [Neo4j]() Topics: [Hackathon](), [Strands Agents](), [Neo4j](), [Graphs](), [Orchestration](), [multimodal-ai](), [OpenAI](), [Open Source]() Tags: [agentic-ai](), [ai-agents-in-action](), [ai-video-model](), [aws](), [context-graph](), [developer](), [graph](), [hackathon](), [hackathons](), [knowledge-graph](), [multimodal](), [neo4j](), [open-source](), [openai](), [orchestration](), [repo](), [strands-agents]() ### AI overview A report on a video-agent hackathon where teams combined AWS Strands Agents, OpenAI, TwelveLabs, and Neo4j. It describes a starter repository that helped teams build video-to-graph-to-agent applications and reviews how the top projects used graph-based context. ### Source excerpt Last Thursday, 151 builders filled the AWS Builder Loft on Market Street for Hack the Video Agent Context Graph, hosted by HackerSquad with the AWS Builder Loft -- one day, one theme, 37 projects, and four required tools: AWS Strands Agents for orchestration, OpenAI for reasoning, TwelveLabs for multimodal video understanding, and Neo4j for the context graph. Several teams, including the winner, used a Neo4j starter repo that showed how all the tools could work together. Here's what happened and why you should use a starter repo for your next hackathon! ## How AI Agents Make Decisions and How Context Graphs Can Improve Their Reliability DevFeed: [How AI Agents Make Decisions and How Context Graphs Can Improve Their Reliability]() Original publisher: [Read original article]() Author: Alyssa Di Pasqualucci Published: 2026-07-20T17:29:46Z Content type: tutorial Language: en Sources: [Neo4j]() Topics: [AI Agent](), [decision-making](), [context](), [Artificial Intelligence]() Tags: [agentic-ai](), [agents](), [ai](), [ai-agents](), [context](), [context-graph](), [decision-making](), [traces]() ### AI overview This guide explains how AI agents make decisions through a loop of understanding goals, gathering context, choosing an action, executing it, and learning. It argues that connected context, including business knowledge, conversation history, and decision traces, can make agent decisions more grounded, explainable, and improvable. ### Source excerpt AI agents could transform enterprise productivity with their ability to handle complex workflows, make decisions, and act autonomously toward business goals. But this promise can't be fulfilled if agents sound confident while getting things wrong, miss critical context, or lose... Read more -> ## A Tour of the Neo4j Agent Memory Service (NAMS) DevFeed: [A Tour of the Neo4j Agent Memory Service (NAMS)]() Original publisher: [Read original article]() Author: Jocelyn Hoppa Published: 2026-06-23T15:19:00Z Content type: tutorial Language: en Sources: [Neo4j]() Topics: [Neo4j](), [Artificial Intelligence](), [graph-database](), [Cloud](), [API](), [Model Context Protocol](), [REST API](), [Large Language Model](), [Database](), [dashboards](), [async]() Tags: [agent](), [agent-memory](), [ai](), [ai-agent](), [ai-agents](), [api](), [async](), [cloud](), [context-graph](), [context-window](), [database](), [developer](), [genai](), [graph-database](), [http](), [llm](), [mcp](), [neo4j](), [neo4j-agent-memory](), [ontology]() ### AI overview A walkthrough of Neo4j Agent Memory Service, an experimental managed cloud service that provides persistent, structured memory for LLM agents. It uses Neo4j Aura, REST or MCP access, entity extraction, deduplication, embeddings, compression, and context retrieval. ### Source excerpt A walkthrough of the graph-native memory layer for AI agents.Most agents you build today are amnesiacs. They run a chain of reasoning, return an answer, and forget the whole thing the moment the request ends. The usual patch of stuffing transcripts... Read more -> ## POLE+O: The 5-Type Ontology That Solves the Hardest Part of Building a Knowledge Graph DevFeed: [POLE+O: The 5-Type Ontology That Solves the Hardest Part of Building a Knowledge Graph]() Original publisher: [Read original article]() Author: Jocelyn Hoppa Published: 2026-06-15T20:08:00Z Content type: article Language: en Sources: [Neo4j]() Topics: [Graphs](), [Neo4j](), [AI Agent](), [data](), [Structured-data](), [Command-line interface](), [context](), [Full Stack Development](), [object]() Tags: [agent](), [ai-agent](), [ai-agent-applications](), [cli](), [context-graph](), [data](), [data-modeling](), [developer](), [graph](), [knowledge-graph](), [neo4j](), [node](), [ontology](), [project](), [schema]() ### AI overview The article explains POLE+O, a five-type ontology for starting knowledge graph projects with Person, Organization, Location, Event, and Object as base entity types. It describes how Neo4j's Create Context Graph uses this schema with multi-label modeling to support domain mapping, querying, and entity unification. ### Source excerpt Neo4j's Create Context Graph ships with a universal starting schema. Here's why starting with 5 node types beats starting from scratch.image from create-context-graph homepage (https://create-context-graph.dev/)The blank canvas problemEvery knowledge graph project starts the same way. Someone opens a whiteboard and asks: "So...... Read more ->