Graphthulhu
MCP server that gives AI full access to your Logseq or Obsidian knowledge graph. 39 tools for navigation, search, analysis, writing, decisions, journals, flashcards, and whiteboards.
What is Graphthulhu?
Graphthulhu is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to mcp server that gives ai full access to your logseq or obsidian knowledge graph. 39 tools for navigation, search, analysis, writing, decisions, journals, flashcards, and whiteboards.
MCP server that gives AI full access to your Logseq or Obsidian knowledge graph. 39 tools for navigation, search, analysis, writing, decisions, journals, flashcards, and whiteboards.
This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- MCP server that gives AI full access to your Logseq or Obsid
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx graphthulhuConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Graphthulhu
Graphthulhu is a Go-based MCP server that gives AI assistants direct access to your Logseq or Obsidian knowledge graph through 37 specialized tools spanning navigation, search, writing, analysis, journaling, flashcards, and whiteboards. It connects to Logseq via its HTTP API or reads Obsidian vaults directly from disk, enabling your AI assistant to read, write, search, and analyze your entire personal knowledge base. Researchers, writers, and knowledge workers use it to let AI assistants actively work within their second brain — finding connections, filling gaps, and creating structured content.
Prerequisites
- Go 1.21+ if building from source, or download the pre-built binary from GitHub Releases
- Logseq desktop app with HTTP APIs enabled (Settings > Features > HTTP APIs), or an Obsidian vault on disk
- A Logseq API token generated from the Logseq HTTP API settings (for Logseq backend)
- The Obsidian vault directory path (for Obsidian backend, no plugins required)
- An MCP-compatible client such as Claude Desktop or Claude Code
Install graphthulhu
Install the binary using Go, or download a pre-built release from the GitHub Releases page and add it to your PATH.
go install github.com/skridlevsky/graphthulhu@latestEnable Logseq HTTP APIs (Logseq users only)
In Logseq, open Settings > Features and enable the HTTP APIs option. Start the HTTP server and copy the generated Bearer token.
Set environment variables for your backend
Configure the backend type and credentials. For Logseq set the API URL and token; for Obsidian set the vault path.
# For Logseq:
export GRAPHTHULHU_BACKEND=logseq
export LOGSEQ_API_URL=http://127.0.0.1:12315
export LOGSEQ_API_TOKEN=your_logseq_bearer_token
# For Obsidian:
export GRAPHTHULHU_BACKEND=obsidian
export OBSIDIAN_VAULT_PATH=/path/to/your/vaultTest the server runs correctly
Run graphthulhu directly to verify it connects to your knowledge graph before wiring it into an MCP client.
graphthulhuAdd graphthulhu to your MCP client configuration
Add the server to your claude_desktop_config.json or Claude Code config, passing environment variables inline.
Graphthulhu Examples
Client configuration
Add this to your claude_desktop_config.json. Update the env block with your actual backend choice and credentials.
{
"mcpServers": {
"graphthulhu": {
"command": "graphthulhu",
"args": [],
"env": {
"GRAPHTHULHU_BACKEND": "logseq",
"LOGSEQ_API_URL": "http://127.0.0.1:12315",
"LOGSEQ_API_TOKEN": "your_logseq_bearer_token"
}
}
}
}Prompts to try
Use these prompts to interact with your Logseq or Obsidian knowledge graph through Claude.
- "Search my knowledge graph for notes about machine learning and summarize the key themes"
- "Find all orphaned pages in my graph that have no incoming links"
- "Create a new page called 'Project Alpha' with a summary and link it to my existing 'Projects' page"
- "Show me all flashcards that are due for review today"
- "What are the topic clusters in my knowledge graph? Identify the top 5 themes"Troubleshooting Graphthulhu
Connection refused error when using Logseq backend
Logseq's HTTP API server must be running. Open Logseq, go to Settings > Features, confirm HTTP APIs is enabled, and verify the server is listening on port 12315. Also make sure LOGSEQ_API_URL matches the port shown in Logseq settings.
Obsidian vault not found or permission denied
Verify that OBSIDIAN_VAULT_PATH points to the exact directory containing your vault's .md files (the root folder, not a parent). On macOS you may need to grant Terminal or your MCP client access to that directory in System Settings > Privacy & Security > Files and Folders.
'graphthulhu: command not found' after go install
Go installs binaries to $(go env GOPATH)/bin. Run 'export PATH=$PATH:$(go env GOPATH)/bin' and add this line to your ~/.zshrc or ~/.bashrc so the command is available in future sessions.
Frequently Asked Questions about Graphthulhu
What is Graphthulhu?
Graphthulhu is a Model Context Protocol (MCP) server that mcp server that gives ai full access to your logseq or obsidian knowledge graph. 39 tools for navigation, search, analysis, writing, decisions, journals, flashcards, and whiteboards. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Graphthulhu?
Follow the installation instructions on the Graphthulhu GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Graphthulhu?
Graphthulhu works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Graphthulhu free to use?
Yes, Graphthulhu is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
Graphthulhu Alternatives — Similar Knowledge & Memory Servers
Looking for alternatives to Graphthulhu? Here are other popular knowledge & memory servers you can use with Claude, Cursor, and VS Code.
MemPalace
★ 52.6kA local AI memory system that stores all conversations verbatim and organizes them into navigable structures. It provides 19 MCP tools for AI assistants to search and retrieve past decisions, debugging sessions, and architecture debates automatically
Kratos
★ 25.7k🏛️ Memory System for AI Coding Tools - Never explain your codebase again. MCP server with perfect project isolation, 95.8% context accuracy, and the Four Pillars Framework.
Context Mode
★ 15.4kAn MCP server that preserves LLM context by intercepting large data outputs and returning only concise summaries or relevant sections. It enables efficient sandboxed code execution, file processing, and documentation indexing across multiple programm
Memu
★ 13.7kMemory for 24/7 proactive agents like OpenClaw.
MemOS
★ 9.3kMemOS (Memory Operating System) is a memory management operating system designed for AI applications. Its goal is: to enable your AI system to have long-term memory like a human, not only remembering what users have said but also actively invoking, u
Everos
★ 5.4kBuild, evaluate, and integrate long-term memory for self-evolving agents.
Browse More Knowledge & Memory MCP Servers
Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Graphthulhu in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Graphthulhu?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.