Graphthulhu is a comprehensive Model Context Protocol server that enables AI agents to read, write, and analyze Logseq and Obsidian knowledge graphs.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install graphthulhu
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 graphthulhu 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 Graphthulhu?
Graphthulhu is a specialized MCP server designed to turn your personal knowledge base into a fully accessible data source for AI agents. As a key utility within the Openclaw Skills ecosystem, it provides a bridge to both Obsidian vaults and Logseq graphs, offering 37 distinct tools across nine categories. This allows developers and knowledge workers to interact with their second brain using natural language, leveraging the power of AI to synthesize information and manage notes.
Whether you are managing a markdown-based Obsidian vault or a block-structured Logseq database, Graphthulhu handles the heavy lifting of graph traversal, full-text search, and bidirectional link management. By integrating this skill, users can transform static documentation into a dynamic, queryable graph that enhances productivity and automated content creation.
Graphthulhu Use Cases
- Automating the creation of complex note structures and nested blocks within a personal knowledge base.
- Performing advanced graph analysis to find topic clusters, knowledge gaps, and orphaned notes.
- Managing academic or professional learning through Spaced Repetition System (SRS) flashcard integration.
- Querying historical journal entries and decision logs to track project progress over time.
- Refactoring large knowledge graphs by renaming pages and automatically updating all related bidirectional links.
How Graphthulhu Works
- The user installs the Graphthulhu binary and registers it as an MCP server within an AI coding agent or environment compatible with Openclaw Skills.
- For Obsidian, the server points directly to the local vault path; for Logseq, it connects via the local HTTP API and a secure token.
- When a user requests information from their notes, the AI agent invokes specific Graphthulhu tools to search, list pages, or fetch block hierarchies.
- Graphthulhu processes the request, traversing the link graph or executing Datalog queries to retrieve relevant context.
- The tool returns structured Markdown or JSON data, which the agent uses to generate insights or perform further write operations like appending blocks.
Graphthulhu Setup
Download the latest binary from the GitHub releases page and add it to your PATH, or install via Go:
go install github.com/skridlevsky/graphthulhu@latest
Obsidian Integration: Add the following configuration to your MCP settings file:
{
"mcpServers": {
"graphthulhu": {
"command": "graphthulhu",
"args": ["--backend", "obsidian", "--vault", "/path/to/your/vault"]
}
}
}
Logseq Integration: Enable the HTTP API server in Logseq (Settings > Features > HTTP APIs), then configure your environment:
{
"mcpServers": {
"graphthulhu": {
"command": "graphthulhu",
"env": {
"LOGSEQ_API_URL": "http://127.0.0.1:12315",
"LOGSEQ_API_TOKEN": "your-token-here"
}
}
}
}
Graphthulhu Data Schema & Taxonomy
Graphthulhu organizes knowledge base data into a schema that balances hierarchical blocks with networked graph connections, making it ideal for Openclaw Skills implementations.
| Component | Description |
|---|---|
| Page | The top-level entity, identified by title, namespace, or tags. |
| Block | The atomic unit of data, supporting nested children and properties. |
| Link | Bidirectional references that form the graph edges between pages. |
| Journals | Date-indexed pages for time-based tracking and searching. |
| Flashcards | Metadata-driven blocks used for SRS (Logseq specific). |
| Whiteboards | Spatial canvas data for visual knowledge mapping. |
Graphthulhu Advanced Features
- Support for raw Datalog queries in Logseq for complex, multi-parameter data extraction.
- Automated bidirectional link updates during page renaming to maintain graph integrity.
- Bulk property updates and block movement across different pages or hierarchies.
- Specialized search tools for tag hierarchies and property-based filtering.
- Real-time status tracking for decision logs with deferral and resolution workflows.
SKILL.md
Loading
Related Openclaw Skills
Google Hotels Search
Automate hotel searches and price comparisons via browser-based navigation and direct URL date encoding.
Ideation
A sophisticated workflow skill that converts messy brain dumps and dictated thoughts into structured, code-ready implementation artifacts.
PHP Full-Stack Developer
A senior-level governance framework for PHP engineering that prioritizes pre-flight analysis, data integrity, and reproducible verification.
Stellar CLI
A powerful interface for AI agents to manage wallets, send XLM, and deploy smart contracts on the Stellar network.
App Store Deployment Guide
A comprehensive technical manual for navigating the Apple App Store deployment process from account creation to final store approval.
Military PCS Optimizer
A specialized tool for service members to calculate PCS entitlements, plan moving timelines, and maximize financial returns during military relocation.
Featured*
