neo4j-labs/create-context-graph

AI agents with graph based reasoning memory, scaffolded in seconds

What it solves

Create Context Graph is a CLI scaffolding tool that allows developers to quickly build full-stack AI applications centered around a domain-specific knowledge graph. It eliminates the manual effort of designing ontologies, setting up graph databases, and building chat interfaces by generating a complete, functional project in minutes.

How it works

The tool uses an interactive wizard or CLI flags to let users choose an industry domain (or describe a custom one) and a preferred AI agent framework. It then uses Jinja2 templates and LLMs to generate a project containing:

  • Backend: A FastAPI server with an AI agent configured for the chosen domain, integrated with Neo4j for graph memory and LiteLLM for flexible model provider selection.
  • Frontend: A Next.js and Chakra UI interface featuring a streaming chat, real-time tool call visualization, and an interactive graph explorer.
  • Data Layer: A Neo4j schema with domain-specific constraints and indexes, along with optional synthetic demo data or real data imported via SaaS connectors (e.g., GitHub, Slack, Jira, Notion).
  • Integration: Optional MCP server configuration for Claude Desktop integration.

Who it’s for

Developers and AI engineers who want to prototype or deploy domain-specific AI agents that leverage structured graph memory rather than simple vector search, and those who want a ready-made full-stack scaffold for GraphRAG applications.

Highlights

  • Multi-Framework Support: Compatible with PydanticAI, LangGraph, CrewAI, OpenAI Agents SDK, and others.
  • Automated Ontology Generation: Can generate a complete industry-specific ontology from a plain-English description.
  • SaaS Data Connectors: Built-in importers for a wide range of services including Google Workspace, Linear, GitHub, and chat exports from Claude AI and ChatGPT.
  • Flexible Memory Backends: Supports both hosted memory via NAMS and self-hosted Neo4j instances.
  • Comprehensive UI: Includes a decision trace viewer, document browser, and real-time graph visualization.

Related

  • Project
    ddemidov/vexcl

    A C++ vector expression template library for OpenCL and CUDA that reduces boilerplate code for GPGPU development.

  • Project
    nuclia/nucliadb

    NucliaDB is a hybrid AI search database that combines vector, full-text, and graph indexes to store and search unstructured data.

  • Project
    EricLBuehler/candle-vllm

    A high-performance inference and serving platform for local LLMs featuring an OpenAI-compatible API, multi-GPU/multi-node support, and aggressive KV cache compression.

  • Project
    asaotomo/FofaMap

    FofaMap is a Python CLI that blends FOFA asset search with an AI‑driven agent. It can run classic FOFA queries, translate natural‑language requests into multiple queries, self‑reflect to improve results, grade evidence, export to XLSX/CSV/JSONL, and optionally generate a Nuclei scan plan that requires manual approval. It also provides a REST API and one‑click integration as a tool for AI coding assistants (Cursor, Claude, LM Studio, etc.).