aws/context-ontology-accelerator

An open-source, ontology-based semantic context accelerator that enables AI agents to make more accurate, consistent, and explainable decisions.

What it solves

This project provides a semantic context layer for AWS that helps AI agents retrieve accurate, business-aware context. It bridges the gap between raw data sources and AI agents by combining knowledge graphs, formal ontologies, and rule-based systems to ensure that the information retrieved is validated against business logic.

How it works

The system uses a "Scan $\rightarrow$ Model $\rightarrow$ Serve" workflow:

  • Scan: Connects to data sources, discovers schemas, and ingests both structured and unstructured documents.
  • Model: Creates and manages ontologies and a unified semantic graph to define how data relates to business metrics and logic.
  • Serve: Uses SPARQL federation (VKG) and a knowledge graph to serve this validated context to AI agents via the Model Context Protocol (MCP).

Who it’s for

It is designed for developers and architects building AI agents on AWS who need their agents to operate with a deep understanding of complex business domains and formal data relationships.

Highlights

  • Semantic Graph: Combines knowledge graphs with formal ontologies for high-precision context retrieval.
  • MCP Integration: Serves context directly to AI agents using the Model Context Protocol.
  • VKG Support: Implements a Virtual Knowledge Graph for querying via SPARQL federation.
  • ** Ontology Induction**: Includes an engine for ontology induction and reasoning (using HermiT/ELK).
  • Enterprise-grade Access: Features namespace-scoped roles and platform-level access control for secure data management.

Related

  • Project
    semantica-agi/semantica

    Semantica is a graph-native infrastructure layer that gives AI systems structured, explainable context. It builds knowledge graphs from enterprise data, records decisions as traceable nodes, and provides deterministic reasoning, provenance, and audit trails — all without requiring an LLM. It's designed for regulated industries needing accountable AI.

  • Project
    alibaba/UnifiedModel

    UModel (Unified Model) is an open‑source, vendor‑neutral runtime that lets enterprises describe data, services, and infrastructure as a semantic graph. It provides model packs, a workspace‑scoped graph, unified query APIs, a web UI, CLI, and SDKs, and includes AgentGateway/MCP so LLM agents can discover, query, and act on the graph for tasks like root‑cause analysis.

  • Dispatch
    Anthropic Effective Context Engineering for AI Agents

    Anthropic introduces context engineering as a strategy for curating the optimal set of tokens to maximize LLM performance and overcome attention scarcity and context rot.

  • Project
    awslabs/mcp

    A collection of Model Context Protocol (MCP) servers that integrate AWS documentation and infrastructure management tools directly into AI coding assistants and chatbots.

  • Project
    IBM/mcp-context-forge

    An open source registry and proxy that federates MCP, A2A, and REST/gRPC APIs into a single endpoint with centralized governance, discovery, and observability.