Elasticflow MCP Server
Enables AI assistants and teams to collaborate in a shared workspace by building and querying tables, writing documents, uploading files, and publishing live dashboards through any MCP-compatible client.
README
Elasticflow MCP Server
One workspace where your AI agents and your team work together.
Connect your AI assistant to Elasticflow and let it manage tables, write documents, upload files, and publish live dashboards — all within your team's shared workspace. Everything your agents produce is instantly available to the whole organization, not buried in a chat thread.
What you can do:
- Build and query structured data — sales pipelines, contract trackers, campaign analytics, inventory logs
- Create rich documents — reports, proposals, legal summaries, meeting notes
- Upload and organize files — contracts, NDAs, presentations, compliance docs
- Publish live dashboards and client portals — no code required
- Share results across your team — every action is visible, every output is reusable
Prerequisites
You need an Elasticflow account to use this MCP server. When you connect for the first time, your AI assistant will open a browser window where you can sign in or create an account.
Supported Platforms
AI Assistants
- ChatGPT (OpenAI)
- Claude Desktop / Claude Code / Claude.ai
- Google Gemini
- Microsoft Copilot Studio
- Amazon Q
AI-Powered Editors
- Cursor
- VS Code / GitHub Copilot
- Windsurf
- Zed
Productivity Tools
- Raycast
- Cline
Works with any client that supports the MCP Streamable HTTP transport.
Installation
Add the Elasticflow MCP server to your client configuration:
Claude Desktop
Go to Settings > Extensions, or add manually to config:
{
"mcpServers": {
"elasticflow": {
"url": "https://mcp.elasticflow.app/mcp"
}
}
}
Claude Code
claude mcp add elasticflow --transport streamable-http https://mcp.elasticflow.app/mcp
Cursor
Add to .cursor/mcp.json:
{
"mcpServers": {
"elasticflow": {
"url": "https://mcp.elasticflow.app/mcp",
"transport": "streamable-http"
}
}
}
VS Code / GitHub Copilot
Add to .vscode/mcp.json:
{
"servers": {
"elasticflow": {
"type": "streamable-http",
"url": "https://mcp.elasticflow.app/mcp"
}
}
}
Authentication
Elasticflow uses OAuth 2.1 with PKCE. When you connect for the first time, your client will open a browser for authentication. No API keys or manual token configuration needed.
Scopes:
| Scope | Description |
|---|---|
workspaces:read |
View your workspaces |
workspaces:write |
Create and modify workspaces |
tables:read |
View table data and schemas |
tables:write |
Create, modify, and delete table data |
documents:read |
Read your documents |
documents:write |
Create and edit documents |
files:read |
View and download files |
files:write |
Upload files |
interfaces:read |
View interface projects |
interfaces:write |
Build and publish interfaces |
MCP Tools
Workspaces (4 tools)
Organize your team's projects, data, and files into separate workspaces.
| Tool | Description |
|---|---|
mcp_workspaces_list |
See all workspaces your team has access to |
mcp_workspaces_get |
Get workspace details — members, settings, usage |
mcp_workspaces_create |
Set up a new workspace for a team, project, or department |
mcp_workspaces_update |
Rename, configure, or adjust workspace settings |
Tables (15 tools)
Store and manage structured business data — CRM pipelines, inventories, project trackers, or any tabular data your team needs.
| Tool | Description |
|---|---|
mcp_tables_list_tables |
Browse all tables in a workspace |
mcp_tables_get_table |
View table structure — columns, types, row count |
mcp_tables_list_column_types |
See available column types (text, number, date, select, etc.) |
mcp_tables_create_table |
Create a new table for tracking leads, orders, tasks, or anything else |
mcp_tables_update_table |
Rename a table or change its settings |
mcp_tables_delete_table |
Remove a table you no longer need |
mcp_tables_create_column |
Add a new field — e.g. "Deal Size", "Due Date", "Status" |
mcp_tables_update_column |
Change a column's name, type, or options |
mcp_tables_delete_column |
Remove a column from the table |
mcp_tables_get_rows |
Query your data with filters — find overdue tasks, high-value deals, etc. |
mcp_tables_create_row |
Add a new record — a lead, an order, a task |
mcp_tables_update_cells |
Edit values — update a deal stage, mark a task complete |
mcp_tables_delete_rows |
Remove records you no longer need |
mcp_tables_link_rows |
Connect related records across tables — link a contact to a deal |
mcp_tables_unlink_rows |
Remove a connection between records |
Documents (12 tools)
Create and manage rich documents — reports, proposals, meeting notes, knowledge base articles.
| Tool | Description |
|---|---|
mcp_documents_list |
Browse all documents in a workspace |
mcp_documents_read |
Read the full content of a document |
mcp_documents_create |
Write a new report, proposal, or brief |
mcp_documents_update |
Change document title or metadata |
mcp_documents_search |
Find documents by keyword — search across all your team's content |
mcp_documents_patch |
Edit a specific section without rewriting the whole document |
mcp_documents_append |
Add new content to the end — meeting notes, status updates |
mcp_documents_insert_image |
Add a chart, screenshot, or diagram to a document |
mcp_documents_insert_file |
Attach a spreadsheet, PDF, or any file to a document |
mcp_documents_list_assets |
See all images and files embedded in a document |
mcp_documents_remove_asset |
Remove an outdated image or attachment |
mcp_documents_replace_asset |
Swap an asset with an updated version |
Files (4 tools)
Upload, organize, and access files — contracts, presentations, exports, media.
| Tool | Description |
|---|---|
mcp_files_list |
Browse uploaded files |
mcp_files_get_metadata |
Check file details — size, type, upload date |
mcp_files_get_content |
Download or read a file's contents |
mcp_files_upload |
Upload a new file — CSV, PDF, image, or any format |
Interfaces (11 tools)
Build and publish live dashboards, client portals, and internal tools — no code required.
| Tool | Description |
|---|---|
mcp_interfaces_list_projects |
See all your dashboards and web apps |
mcp_interfaces_get_project |
View project configuration and settings |
mcp_interfaces_get_build_status |
Check if your latest changes are deployed |
mcp_interfaces_get_build_logs |
Debug build issues by reading the logs |
mcp_interfaces_discover_schema |
Auto-detect data sources available for your dashboard |
mcp_interfaces_create_project |
Start a new dashboard or client portal |
mcp_interfaces_update_project |
Change layout, data sources, or design |
mcp_interfaces_configure_access |
Control who can view — share with clients or lock to your team |
mcp_interfaces_build |
Compile your changes and prepare for deployment |
mcp_interfaces_publish |
Push your dashboard live for viewers |
mcp_interfaces_rollback |
Something broke? Instantly revert to the previous version |
Install Only What You Need
The main server (/mcp) gives you all 51 tools at once. If you only need part of the platform, or your AI client has a tool limit, you can install smaller servers instead:
| Server | URL | Tools | What you get |
|---|---|---|---|
| All | https://mcp.elasticflow.app/mcp |
51 | Everything |
| Data | https://mcp.elasticflow.app/data |
19 | Tables, rows, columns, workspaces |
| Content | https://mcp.elasticflow.app/content |
16 | Documents, search, files, uploads |
| Platform | https://mcp.elasticflow.app/platform |
11 | Dashboards, portals, publishing |
Each server uses the same account and authentication — install one or all three, they work independently.
Example: install only Data and Content (skip dashboards)
Claude Code:
claude mcp add elasticflow-data --transport streamable-http https://mcp.elasticflow.app/data
claude mcp add elasticflow-content --transport streamable-http https://mcp.elasticflow.app/content
Cursor (.cursor/mcp.json):
{
"mcpServers": {
"elasticflow-data": {
"url": "https://mcp.elasticflow.app/data",
"transport": "streamable-http"
},
"elasticflow-content": {
"url": "https://mcp.elasticflow.app/content",
"transport": "streamable-http"
}
}
}
Usage Examples
Sales
"Create a table called Pipeline with columns: company, deal size, stage, close date, owner"
"Show me all deals closing this month sorted by deal size"
"Enrich leads in the Prospects table — fill in company size, industry, and LinkedIn profile"
"Prepare objection responses for the top 5 deals stuck in negotiation stage"
"Create a document with a weekly pipeline summary from the Pipeline table"
"Build a client-facing dashboard showing project milestones for Acme Corp"
Legal
"Create a table to track active contracts: vendor, type, start date, expiry, auto-renewal"
"Search documents for all mentions of indemnification clause"
"List all contracts expiring in the next 90 days"
"Upload the signed NDA and link it to the Vendors table"
Marketing
"Create a campaign tracker table with columns: campaign, channel, budget, spend, leads, CPL"
"Create a document with a monthly marketing report and insert the performance chart"
"Show me all campaigns where CPL is above $50"
"Publish the campaign results dashboard for the leadership team"
Operations
"Create a workspace for Q2 operations and set up tables for inventory, suppliers, and orders"
"Show me all suppliers with overdue deliveries from the Orders table"
"Create a document summarizing this week's operational incidents"
"Build and publish a live dashboard tracking order fulfillment rates"
Documentation
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。