Roam Research

Roam Research is an MCP server that connects Claude and other Model Context Protocol clients directly to your Roam Research graph. Built for researchers, knowledge workers, and personal note-takers, the server allows AI assistants to read, query, and modify graph data through plain conversational prompts without custom scripts. Users can create nested outlines, add daily notes, manage TODO items, and modify block hierarchies while preserving Roam native structure. In addition to basic block and page operations, the server supports complex Datalog queries, block reference traversals, and a dedicated memory system that lets an assistant store and recall categorized facts across chats. It also includes utilities to parse content from URLs, PDF files, and YouTube transcripts before inserting summaries or raw text into designated graph locations. By exposing the Roam API as standardized MCP tools, it enables language models to act as interactive knowledge base operators.

Category: Notes & Knowledge Management

Tags: graph database, knowledge-management, notes, roam research

Visit Roam Research

How to install and configure Roam Research

To install and configure the Roam Research MCP server for Claude Desktop: 1. Open your Claude Desktop configuration file: - macOS: ~/Library/Application Support/Claude/claude_desktop_config.json - Windows: %APPDATA%\Claude\claude_desktop_config.json 2. Obtain an API token in your Roam Research graph settings under "API tokens" and note your graph name. 3. Add the server definition inside the mcpServers object: json { "mcpServers": { "roam-helper": { "command": "uvx", "args": ["git+https://github.com/PhiloSolares/roam-mcp.git"], "env": { "ROAM_API_TOKEN": "<your_roam_api_token>", "ROAM_GRAPH_NAME": "<your_roam_graph_name>" } } } } 4. Restart Claude Desktop to apply the configuration. Alternatively, build and run the Docker container described in the repository README.

What you can do with Roam Research

• Creating structured meeting notes or project pages with indented block hierarchies and automatic TODO tagging. - Searching graph databases for specific tags, nested block references, or items filtered by creation and modification dates. - Running custom Datalog queries to locate interconnected nodes across large, complex personal knowledge graphs. - Maintaining persistent cross-session memories by storing tagged user notes under dedicated graph pages like Memories. - Extracting text from external URLs, PDFs, or YouTube transcripts and appending summaries directly into daily notes.

Key facts

• https://github.com/PhiloSolares/roam-mcp
• Notes & Knowledge Management
• graph database, knowledge-management, notes, roam research

Part of MCP Servers

What can Roam Research MCP server do?

The server enables AI assistants to create pages, insert structured blocks, update existing content, search notes by tag or modification date, and run advanced Datalog queries. It also provides tools to store persistent assistant memories and parse external content from PDFs, web pages, and YouTube video transcripts.

How do I install Roam Research MCP?

You configure Claude Desktop by editing its JSON configuration file to run the server via uvx pointing to git+https://github.com/PhiloSolares/roam-mcp.git. You must provide your Roam API token and graph name as environment variables, or run the server in a pre-built Docker container.

Which MCP clients work with Roam Research?

The server works with Claude Desktop and any client that implements the Model Context Protocol standard and supports local command execution or standard containerized MCP server integrations.

Is Roam Research MCP open source?

Yes, Roam Research MCP server is open source software published under the MIT License on GitHub.