Verified:GitHub

GraphRAG Knowledge Graph

Ingest raw text into a GraphRAG knowledge graph and run graph-aware queries.

Updated Apr 11, 2026

Search Tags:#knowledge graph #semantic search #document ingestion #relationship management #entity extraction #graphrag #workspace indexing

Ready to run

npx skills add https://github.com/sharpninja/mcpserver-copilot-plugin --skill graphrag-knowledge-graph

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: GraphRAG Knowledge Graph
Source: https://github.com/sharpninja/mcpserver-copilot-plugin/tree/main/skills/graphrag
Command: npx skills add https://github.com/sharpninja/mcpserver-copilot-plugin --skill graphrag-knowledge-graph

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

GraphRAG helps teams turn scattered text into a queryable knowledge graph so answers can be grounded in entities and relationships rather than only keyword matches.

Core Features & Use Cases

Ingest raw text into a GraphRAG corpus: Add ad-hoc content as documents and control whether a full reindex is triggered, useful for incremental knowledge updates.
Index and query a knowledge graph: Verify GraphRAG status, rebuild the index, and run natural-language graph queries in local/global/drift modes for different retrieval behaviors.
Manage knowledge graph structure: Create, list, update, and delete entities and relationships, enabling curated concepts like systems, components, and how they interact; use cases include building maintainable AI-assisted support and architecture maps.

Quick Start

Ask the skill to ingest your content by providing the text, title, and a source key, then query it once indexing is complete.

Frequently Asked Questions about GraphRAG Knowledge Graph

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

FAQPage Schema

How do I ingest raw text into a knowledge graph for semantic search?▼

You ingest raw text into a knowledge graph by providing the content, title, and source key to add it as a document, optionally triggering a full reindex for incremental knowledge updates.

How does GraphRAG querying differ from standard keyword search?▼

GraphRAG querying grounds answers in extracted entities and relationships rather than keyword matches, enabling local, global, and drift retrieval modes over the indexed workspace graph.

What is the best way to manage entities and relationships in a knowledge graph?▼

Manage knowledge graph entities and relationships by creating, listing, updating, and deleting typed connections to curate concepts like systems and components for AI-assisted architecture maps.

Do I need to verify GraphRAG status before running graph-aware queries?▼

Yes, running graph-aware queries requires mandatory GraphRAG status checks to verify the workspace index is built before executing local, global, or drift retrieval operations.

Can I update my knowledge graph corpus without triggering a full reindex?▼

Yes, you can add ad-hoc content as documents without a full reindex by using the controlled ingestion trigger parameter, allowing incremental knowledge updates to the corpus.