graphiti
Build Real-Time Knowledge Graphs for AI Agents
Graphiti is an MCP server that enables AI agents to build and query temporal knowledge graphs that evolve over time. Unlike traditional knowledge graphs, Graphiti tracks how facts change historically, maintains provenance to source data, and supports both prescribed and learned ontologies. The server provides episode management, entity and relationship handling, semantic search, and graph maintenance operations through the Model Context Protocol, making it ideal for context-aware AI applications that need to remember and reason about evolving information.
Key Features
Use Cases
- 01Building AI agents that need to track customer preferences and behaviors as they evolve over time
- 02Creating conversational assistants that remember past interactions with full temporal context
- 03Developing enterprise AI applications that integrate structured and unstructured data into queryable graphs
- 04Implementing personalized recommendation systems that adapt to changing user interests
- 05Building knowledge management systems that preserve historical context while staying current
- 06Creating AI agents that can answer 'what was true then' versus 'what is true now' queries
Related MCP Servers
View moremarkitdown
Python tool for converting files and office documents to Markdown.
firecrawl
The context API to search, scrape, and interact with the web at scale. 🔥
prompts.chat
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
langflow
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
graphiti — FAQ
What is the Graphiti MCP server?+
The Graphiti MCP server is a Model Context Protocol implementation that allows AI assistants to build and query temporal knowledge graphs. It provides episode management, entity and relationship handling, semantic search, and graph maintenance capabilities through the MCP protocol.
How do I install the Graphiti MCP server?+
Install Graphiti with 'pip install graphiti-core' or 'uv add graphiti-core'. You'll also need Python 3.10+, a graph database (Neo4j 5.26+, FalkorDB 1.1.2+, or Amazon Neptune), and an OpenAI API key (or alternative LLM provider). Optional extras are available for specific backends like [falkordb], [neptune], or LLM providers like [anthropic].
What API keys or credentials does Graphiti require?+
Graphiti requires an OpenAI API key by default for LLM inference and embeddings. It also supports Anthropic, Google Gemini, Groq, Azure OpenAI, and OpenAI-compatible local LLMs (Ollama, vLLM, etc.). You'll also need credentials for your chosen graph database backend (Neo4j, FalkorDB, or Amazon Neptune).
Which AI clients work with the Graphiti MCP server?+
The Graphiti MCP server works with any AI assistant that supports the Model Context Protocol. The README mentions integration with AI assistant workflows through MCP, and it can be deployed using Docker with Neo4j.
Is Graphiti free to use?+
Yes, Graphiti is open-source and free to use. However, you'll need to pay for LLM API usage (OpenAI, Anthropic, etc.) and potentially for your graph database hosting, depending on your chosen backend and deployment method.
What graph databases does Graphiti support?+
Graphiti supports Neo4j 5.26+, FalkorDB 1.1.2+ (including embedded FalkorDB Lite), Amazon Neptune Database Cluster and Neptune Analytics Graph, and Kuzu 0.11.2 (deprecated). Each backend requires specific installation extras when installing graphiti-core.
How do I install graphiti?+
Open the source repository on GitHub and follow its README. graphiti is a mcp server — MCP Agents Market links you directly to the official repo.
Is graphiti free?+
graphiti is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.