A robust interface for managing knowledge graph facts, episodes, and entity relationships via the Graphiti API.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install graphiti
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install graphiti using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Graphiti Knowledge Graph?
Graphiti provides a sophisticated infrastructure for building and querying knowledge graphs, enabling AI agents to maintain persistent, high-fidelity memory. By leveraging Openclaw Skills, developers can integrate fact extraction and relationship mapping directly into their agentic workflows. This skill bridges the gap between raw data and structured intelligence by connecting to a Graphiti service supported by Neo4j for graph storage and Qdrant for vector search. Openclaw Skills facilitate a modular approach to building AI-driven knowledge bases that evolve with every interaction.
Graphiti Knowledge Graph Use Cases
- Retrieving specific facts from a knowledge graph to augment LLM context during complex tasks.
- Storing new conversational episodes or project milestones as persistent memories for future recall.
- Automatically extracting entities and relationships from unstructured text to build a structured database.
- Building long-term memory systems for autonomous AI agents that require historical context.
How Graphiti Knowledge Graph Works
- The skill performs a dynamic discovery check to locate the active Graphiti service URL using standard Openclaw Skills environment protocols.
- When a search is initiated, the tool sends a vectorized query to the Graphiti facts search endpoint.
- The backend utilizes Qdrant and Neo4j to identify and return the most relevant knowledge nodes based on the query context.
- When adding data, the skill posts content to the messages endpoint for ingestion.
- Graphiti processes the incoming content, extracts entities, and updates the graph schema, making new facts immediately searchable.
Graphiti Knowledge Graph Setup
To begin using this skill within Openclaw Skills, ensure you have a running Graphiti stack, including Neo4j and Qdrant. You can deploy the stack using the provided docker-compose configuration.
# Set the Graphiti URL environment variable
export GRAPHITI_URL="http://localhost:8001"
# Alternatively, use the Clawdbot CLI for configuration
clawdbot config set skills.graphiti.baseUrl "http://localhost:8001"
Graphiti Knowledge Graph Data Schema & Taxonomy
| Key | Type | Description |
|---|---|---|
| query | string | The natural language search term for fact retrieval. |
| max_facts | integer | The maximum number of facts to return in a single search. |
| name | string | The identifier or title for a new memory episode. |
| content | string | The raw text content to be ingested into the graph. |
Graphiti Knowledge Graph Advanced Features
- Automated environment discovery via internal reference scripts within the Openclaw Skills ecosystem.
- Native support for vector-based fact search and retrieval for semantic accuracy.
- Seamless configuration management using the Clawdbot CLI to handle base URLs.
- Extensible architecture allowing for multi-service graph orchestration across different database backends.
SKILL.md
Loading
Related Openclaw Skills
AgentFin Virtual Card Agent
A powerful API-driven skill for issuing and managing USDT-funded virtual payment cards for automated purchasing workflows.
Prompt Defense (Email)
A specialized security layer designed to detect and neutralize prompt injection attacks within email communications.
BotBrag
A visibility platform for AI agents to rank on a global Bitcoin-based leaderboard through satoshi donations.
Amap Maps Skill
A powerful MCP skill for integrating Amap LBS services like geocoding, route planning, and POI search into AI agent workflows.
Failover Gateway
Automate high availability for your AI agent with an active-passive failover system that promotes a standby node when your primary instance fails.
exo-installer
The centralized command-line manager for installing, updating, and health-checking the entire E.x.O. suite of AI agent tools.
Featured*
