{"slug": "give-any-ai-assistant-direct-access-to-your-knowledge-graph-with-mcp", "title": "Give Any AI Assistant Direct Access to Your Knowledge Graph with MCP", "summary": "Chaos Cypher has added support for the Model Context Protocol (MCP), Anthropic's open standard for connecting AI assistants to external tools, allowing assistants like Claude Desktop, Claude Code, Cursor, and Windsurf to directly query, search, traverse, and write to a user's knowledge graph. The integration eliminates manual copy-paste workflows by exposing graph operations such as graphrag_search, find_shortest_path, search_nodes, and get_node_context as callable tools, letting AI hosts retrieve grounded answers from indexed entities, relationships, and source documents in a single turn.", "body_md": "Your knowledge graph is stuck in a browser tab. You built something valuable -- a map of entities, relationships, and source documents that represents real understanding of a domain. But the moment you switch to Claude to write a report, or open Cursor to write code, or ask ChatGPT to help with analysis, that knowledge graph might as well not exist. You're back to copying text, pasting context, and manually cross-referencing. Two tools that should be working together are stuck in separate worlds.\n\nChaos Cypher now speaks MCP, which means any AI assistant that supports the protocol -- Claude Desktop, Claude Code, Cursor, Windsurf, and a growing list of others -- can directly query, search, traverse, and even write to your knowledge graph. No copy-paste. No context switching. Just ask.\n\nThis post walks through what that actually looks like, what's under the hood, and how to set it up in about two minutes.\n\nMCP stands for Model Context Protocol. Anthropic released it as an open standard, and the simplest analogy is USB-C for AI tools. Before USB-C, every device had its own charger, its own cable, its own connector. MCP does the same thing for AI integrations: it defines one protocol that any AI host can use to talk to any tool server.\n\nInstead of building a custom plugin for Claude, another for ChatGPT, another for Cursor, and another for every new AI tool that launches next month, you build one MCP server. Every compatible AI tool can use it immediately.\n\nThe adoption has been fast. Claude Desktop, Claude Code, Cursor, Windsurf, Cline, and Continue all support MCP today. The protocol handles tool discovery (the AI asks \"what can you do?\"), tool invocation (the AI calls a function with parameters), and result streaming. From the AI's perspective, your knowledge graph becomes just another set of capabilities it can use to answer questions.\n\nFrom your perspective, it means you stop being the middleman between your data and your AI.\n\nThe best way to understand MCP is to see the before and after.\n\n**Before MCP:** You have a knowledge graph with 200 entities extracted from research papers on gene therapy. You're writing a literature review in Claude. To reference your graph, you open Chaos Cypher in another tab, run a search, copy the results, paste them into Claude, ask your question, realize you need more context, go back to the graph, find related entities, copy those too, paste again. Repeat until frustrated.\n\n**After MCP:** You tell Claude: \"Search my knowledge graph for all entities related to CRISPR and find the shortest path to gene therapy applications.\" Claude calls `graphrag_search` to find relevant entities and document passages, then calls `find_shortest_path` to trace the relationship chain. You get a grounded answer with specific entities and relationships from your own research, in one turn.\n\nHere are three scenarios that show the range of what's possible.\n\nYou've been building a knowledge graph from papers on quantum computing and machine learning. You're deep in a writing session in Claude Desktop and want to understand where these two fields intersect in your collected research.\n\nYou ask: *\"What are the connections between quantum computing and machine learning in my research? Show me the key entities and how they're related.\"*\n\nClaude calls `search_nodes` to find nodes matching both topics, then `get_node_context` to pull the immediate neighborhood of the most central ones, including the edges that connect them and the source document chunks that support each relationship. You get back a structured map of how your research connects these fields -- not a generic internet answer, but one grounded in the specific papers you've indexed.\n\nYou're in Cursor, working on a codebase that has an associated knowledge graph mapping its architecture -- services, APIs, data flows, dependencies. You need to understand how the authentication service connects to the billing pipeline.\n\nYou ask: *\"Traverse from the Authentication Service node to anything related to billing. What's the path?\"*\n\nCursor calls `resolve_node` to find the canonical node for \"Authentication Service\" (even if you didn't remember the exact label), then `traverse_path` to walk the graph two hops out, filtered to the relevant edge types. You see the chain: Authentication Service -> User Session -> Subscription Manager -> Billing Pipeline. Without leaving your editor.\n\nYou're drafting a report and need to summarize everything in your knowledge graph about a specific topic, with citations back to the original source documents.\n\nYou ask: *\"Summarize all my sources related to climate policy in the European Union. Include which documents each claim comes from.\"*\n\nClaude calls `get_summary_context` to retrieve and cluster document chunks relevant to the query. Because this tool returns the raw chunks with their source metadata rather than making an LLM call, Claude itself does the summarization -- giving you a synthesis grounded in your documents, with each claim traced back to a specific source.\n\nChaos Cypher exposes 31 tools through MCP, organized into seven categories: **GraphRAG search** (the flagship -- Personalized PageRank fused with hybrid vector/keyword retrieval), **nodes**, **edges**, **templates**, **analytics** (shortest paths, similarity, community detection, multi-hop traversal), **documents** (upload, status, summarization context), and **client-driven extraction** -- where the AI assistant reads chunks, extracts entities itself, and submits them back, no server LLM required. The [full tool reference is in the docs](https://chaoscypher.com/docs/user-guide/mcp#available-tools).\n\nThe design principle: read operations are always safe and always available, write operations are opt-in. **19 tools are read-only; 12 require write mode to be explicitly enabled** by a single setting. If you're not comfortable with an AI modifying your graph, leave it in read mode -- the AI can still search, traverse, and analyze everything.\n\n**Two transport modes:** The MCP server runs in two ways depending on your setup:\n\n`/api/v1/mcp` using the Streamable HTTP transport, so MCP clients on your network can connect after authenticating.\nBoth transports expose the same 31 tools with the same behavior. The only difference is how they're connected.\n\nIf you have the Chaos Cypher CLI installed, connecting Claude Code is one command:\n\n```\nclaude mcp add chaoscypher -- chaoscypher mcp\n```\n\nFor Claude Desktop, add the equivalent two-line entry to `claude_desktop_config.json`:\n\n```\n{\n  \"mcpServers\": {\n    \"chaoscypher\": { \"command\": \"chaoscypher\", \"args\": [\"mcp\"] }\n  }\n}\n```\n\nThat's genuinely it -- restart the client and Chaos Cypher's tools appear automatically. Cursor and other stdio clients use the same two-line config, and if you run the Docker stack, the MCP endpoint is *already live* at `http://localhost/api/v1/mcp` for any Streamable-HTTP client that signs in with your Chaos Cypher credentials -- it sits behind the same edge authentication as the rest of the API (or `http://localhost:8080/api/v1/mcp` if you run the multi-container dev stack, which publishes Cortex's port directly). Per-client walkthroughs are in the [MCP setup docs](https://chaoscypher.com/docs/user-guide/mcp#setup).\n\nBy default the server runs in read-only mode; flip `mcp.mode: write` in `settings.yaml` to enable the 12 write tools, and use `chaoscypher mcp --database my-research` to point at a specific database. The [configuration reference](https://chaoscypher.com/docs/user-guide/mcp#configuration) covers the rest, including `auto_extract` for documents uploaded via MCP.\n\nThis is worth stating explicitly: MCP doesn't send your knowledge graph data to any external service. The protocol is a local communication channel between the AI tool running on your machine and the Chaos Cypher server running on your machine (or your network, if you use Docker). When Claude calls `graphrag_search`, the query goes from Claude to your local MCP server, your server searches your local database, and the results go back to Claude. Your documents, entities, and relationships never leave your infrastructure.\n\nThe AI model itself runs wherever it runs -- that's between you and your provider. But the knowledge graph data stays entirely under your control. If you pair Chaos Cypher with a local model via Ollama, the entire pipeline is air-gapped. See our [local AI setup guide](https://chaoscypher.com/blog/local-ai-knowledge-graph) for the full walkthrough.\n\nMCP support is the foundation for a broader vision: your knowledge graph as a persistent layer that any tool in your workflow can tap into. Here's what's on the roadmap:\n\nThe flagship `graphrag_search` tool deserves its own explanation -- it's doing a lot more than keyword lookup. Read [how GraphRAG works](https://chaoscypher.com/blog/graphrag-enhanced-search) for the full deep-dive on the retrieval pipeline.\n\nThe MCP server ships with Chaos Cypher today. If you're already running it, you have it -- just configure your AI tool and go.\n\n`chaoscypher_core.mcp` package.\nThe gap between \"having a knowledge graph\" and \"using a knowledge graph\" has always been the friction of switching contexts. MCP closes that gap. Your knowledge graph is no longer a destination you visit -- it's a capability that follows you into whatever tool you're already working in.", "url": "https://wpnews.pro/news/give-any-ai-assistant-direct-access-to-your-knowledge-graph-with-mcp", "canonical_source": "https://dev.to/chaoscypherinc/give-any-ai-assistant-direct-access-to-your-knowledge-graph-with-mcp-18b2", "published_at": "2026-09-11 14:45:42+00:00", "updated_at": "2026-09-11 15:14:51.248521+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "ai-products", "developer-tools", "artificial-intelligence"], "entities": ["Chaos Cypher", "Model Context Protocol", "Anthropic", "Claude Desktop", "Claude Code", "Cursor", "Windsurf", "Cline"], "alternates": {"html": "https://wpnews.pro/news/give-any-ai-assistant-direct-access-to-your-knowledge-graph-with-mcp", "markdown": "https://wpnews.pro/news/give-any-ai-assistant-direct-access-to-your-knowledge-graph-with-mcp.md", "text": "https://wpnews.pro/news/give-any-ai-assistant-direct-access-to-your-knowledge-graph-with-mcp.txt", "jsonld": "https://wpnews.pro/news/give-any-ai-assistant-direct-access-to-your-knowledge-graph-with-mcp.jsonld"}}