graphiti-mcp-usage

Guide Graphiti MCP tool usage to capture and retrieve ScreenGraph knowledge.

1|Updated Oct 23, 2025
One-click install
npx skills add https://github.com/nirukk52/ScreenGraph --skill graphiti-mcp-usage
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: graphiti-mcp-usage
Source: https://github.com/nirukk52/ScreenGraph/tree/main/.claude-skills/graphiti-mcp-usage_skill
Command: npx skills add https://github.com/nirukk52/ScreenGraph --skill graphiti-mcp-usage

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

A workflow for using Graphiti MCP tools to capture preferences, procedures, and facts consistently, ensuring knowledge graph hygiene and reuse.

Core Features & Use Cases

  • Before you start, search for Preferences or Procedures and review matches
  • Capture new requirements with add_episode immediately
  • Document procedures and facts for future reuse
  • Tags help organize and retrieve knowledge quickly

Quick Start

  • Load context from Graphiti: search for Preferences/Procedures
  • Add new requirements with add_episode when they arise
  • Structure outputs as Procedures (workflows) and Facts (relationships)

Frequently Asked Questions about graphiti-mcp-usage

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I capture and organize knowledge in a knowledge graph using MCP tools?

Graphiti MCP tools capture knowledge by structuring entries as Preferences, Procedures, and Facts within a screengraph. Search existing entries first, then use add_episode to record new requirements, tagging each entry for efficient retrieval and reuse across your workflow.

What's the difference between Procedures and Facts in a knowledge graph?

Procedures document workflows and processes you follow repeatedly; Facts capture relationships and data attributes. Structuring knowledge this way ensures consistency and makes both searchable and reusable through the MCP search_nodes and search_facts tools.

How do I search for existing Preferences and Procedures before adding new ones?

Use search_nodes within the screengraph group_id to find existing Preferences or Procedures matching your task. Reviewing matches before capturing new requirements prevents duplication and lets you reuse documented workflows.

How should I tag entries in a knowledge graph for future retrieval?

Tag every entry added via add_episode with meaningful labels that describe the preference, procedure, or fact. Consistent tagging ensures you can quickly locate and retrieve related knowledge later using search_nodes across your screengraph scope.

Can I split long requirements when capturing them in a knowledge graph?

Yes. When adding complex requirements with add_episode, break them into smaller, focused entries—each capturing a distinct Preference, Procedure, or Fact. This improves clarity, reusability, and makes updates and searches more efficient.

How do I mark updates to existing knowledge graph entries?

Use add_episode to record updates as new entries within the screengraph, applying tags that link them to previous versions. This maintains an audit trail while keeping your knowledge graph current and organized.