The shared destination is the point: tables, documents and dashboards the assistant creates sit in a team workspace anyone can open. It matters most when the output is meant to be reused rather than read once.
Elasticflow's own remote MCP server, reached over Streamable HTTP at https://mcp.elasticflow.app/mcp. It puts a team workspace behind the client — structured tables, documents, uploaded files and published dashboards — with workspace tools for creating, reading, listing and updating them. What the assistant produces lands in the shared workspace rather than in a chat thread.
- Workspaces created, listed, read and updated from the client
- Structured tables for pipelines, contract trackers, campaign analytics and inventory logs
- Rich documents — reports, proposals, legal summaries, meeting notes
- Files uploaded and organized alongside the data
- Live dashboards and client portals published without code
- Output visible to the whole organization, not only to the person who asked
An Elasticflow account. On first connect the assistant opens a browser window to sign in or register. Any client that speaks the MCP Streamable HTTP transport can connect — there is nothing to install locally.
One command — claude mcp add elasticflow --transport streamable-http https://mcp.elasticflow.app/mcp