This Roam Research MCP server, developed by Ian Shen / 2B3 PRODUCTIONS LLC, provides AI assistants with a standardized interface to interact with Roam Research graphs. Built using TypeScript and leveraging the Roam Research API SDK, it offers capabilities for searching blocks, creating and manipulating pages and blocks, and performing batch operations. The server implements recursive block reference resolution and markdown conversion for content retrieval. It's designed for flexibility, supporting various Roam Research operations through a set of eleven core tools. This implementation is particularly valuable for researchers, writers, and knowledge workers, enabling use cases such as automated note-taking, content organization, and intelligent information retrieval within Roam Research graphs.
Fetch and read a page's content by title, recursively resolving block references up to 4 levels deep.
Create new pages with optional content.
Create new blocks in a page (defaults to today's daily page).
Import nested markdown content under specific blocks.
Add multiple todo items to today's daily page with checkbox syntax.
Create hierarchical outlines with proper nesting and structure.
Search for block references within pages or across the graph.
Navigate and search through block parent-child relationships.
Find all pages that have been modified since midnight today.
Search for blocks containing specific text across all pages or within a specific page.
Update block content with direct text or pattern-based transformations.
Search for blocks and pages based on creation or modification dates.
Search for blocks containing specific tags with optional filtering by nearby tags.
Store and categorize memories or information with automatic tagging.
Recall memories of blocks marked with tag MEMORIES_TAG or blocks on page title of the same name.
Execute custom Datalog queries on the Roam graph for advanced data retrieval and analysis.
Navigate and search through block parent-child relationships.
No reviews yet. Be the first to review!
Sign in to join the conversation
Start the server with node to access it via any client or IDE.
node path/to/downloaded/file.mjs