Remnus
@Ranork
MCP-native open-source Notion alternative. Read and write pages, customizable databases, and kanban/calendar boards through a remote MCP server with OAuth 2.1 and a full audit log.
Overview
What is Remnus?
Remnus is an open-source, MCP-native workspace that serves as an alternative to Notion. It provides pages and customizable databases with table, kanban, and calendar views, and exposes a remote MCP server so AI agents can read and write content directly.
How to use Remnus?
Add the remote endpoint to your MCP client configuration using the URL https://www.remnus.com/api/mcp. Authentication is handled via OAuth 2.1 + PKCE – no token to paste. Once connected, agents can use 14 tools to read and write workspace content.
Key features of Remnus
- Remote hosted MCP server using streamable HTTP
- OAuth 2.1 + PKCE with scoped read/write tokens
- Every agent write is recorded in an audit log
- Pages, databases, kanban, and calendar views
- 14 tools covering read and write operations
- Listed on MCP Registry, Glama, and Glama
Use cases of Remnus
- AI agents searching and querying workspace content
- Agents creating and updating pages dynamically
- Agents managing database schemas and data (table, kanban, calendar)
- Auditing all AI agent writes via the audit log
- Building custom AI-driven knowledge bases with structured data
FAQ from Remnus
How does authentication work?
Authentication uses OAuth 2.1 with PKCE (Proof Key for Code Exchange). No token needs to be pasted – the flow is handled automatically by the MCP client.
What kind of views are supported?
Databases support table, kanban, and calendar views, similar to Notion.
Is Remnus self-hosted or cloud?
Remnus is a remote hosted MCP server. The server URL is https://www.remnus.com/api/mcp.
Can agents write to the workspace?
Yes, agents have access to write tools: create_page, update_page, bulk_update_pages, delete_page, move_item, create_database, and update_database_schema.
Is there an audit log for agent actions?
Yes, every write operation by an agent is recorded in an audit log, which can be queried using the query_audit_log tool.
