Semiont: A Graph Based, AI Native Wiki and Annotator
summary
Semiont is an open source, Apache 2 licensed AI-native wiki that stores document annotations as durable graph and git-backed events, using W3C Web Annotation as the native model and a Neo4j graph plus a Quadrant vector store for retrieval. It runs in containers, supports Llama/Gemma or Anthropic inference, 29 languages, image/PDF assets, and seven intented event verbs (gather, match, bind, yield, mark, browse, beckin). Demo shows unresolved references auto‑generated and committed into the repo.
AI-written summary · about
chapters (18)
- 0:00Introduction
- 0:56Outline and objectives
- 1:29Live demo: wiki interface
- 2:37Resource generation
- 3:45Motivation and core questions
- 5:36Document-grounded knowledge
- 6:39Architectural philosophy
- 7:27Knowledge base use cases
- 9:03Development methodology
- 10:33Installation instructions
- 11:45System configuration
- 13:26Execution and infrastructure
- 14:19Visualizing graphs
- 15:40Core workflow verbs
- 16:57Advanced annotation flows
- 24:25Resolving references
- 28:04System architecture summary
- 29:03Roadmap and conclusion
related talks
46mAn Introduction to Semaphore: Solving Problems with Semantics
25mReduce AI Agents Costs with Semantic tool Selection
33mHow to enhance your own AI services using Semantic Kernel
29mAgentic GraphRAG: Multi-Agent Knowledge Graph Construction for Research Teams
28mAgentic GraphRAG: Multi-Agent Knowledge Graph
36mGENAI-Powered chatbot to increase team efficiency
46mBuilding AI solutions with Semantic Kernel
40mContext That Dreams