Free to use during beta

Connect your AI.Start building with your data.

DataUnmess.ai gives Codex, Claude, Gemini, ChatGPT, Cursor, and Windsurf a secure MCP endpoint for live dashboards, data catalog memory, and workspace artifacts. No new AI subscription. No DataUnmess AI tokens.

Try free: create MCP key See the 3 steps

First run

From blank workspace to first dashboard

1Create a free DataUnmess MCP key

2Copy one prompt into your AI agent

3Ask DataUnmess.ai MCP what to build first

The technical config stays available, but first-time users only need the buttons below.

Guided onboarding

Three steps, no config hunting.

The page keeps the long prompt hidden. Users copy a short action, paste it into their AI agent, then verify by asking DataUnmess.ai MCP to work with real data.

Login and create your MCP key

Open DataUnmess in a new tab, sign in, and create a workspace MCP key. The key is free during beta and shown once.

Copy one prompt into your AI

Pick your client. The copied prompt asks the agent to install DataUnmess MCP with a placeholder, so your real key never goes into chat.

Codex Desktop

Best for this walkthrough. Copies the Codex-specific installer prompt. Codex edits ~/.codex/config.toml and leaves a safe key placeholder.

Preview prompt

I already have a DataUnmess MCP key. Install the DataUnmess MCP server in Codex Desktop by editing this client's MCP config.

Use this endpoint: https://mcp-dev.dataunmess.ai/mcp

For Codex Desktop, add this TOML block to ~/.codex/config.toml (Windows: %USERPROFILE%\.codex\config.toml):

[mcp_servers.dataunmess]
url = "https://mcp-dev.dataunmess.ai/mcp"
http_headers = { Authorization = "Bearer <YOUR_MCP_KEY>" }

Keep the literal string <YOUR_MCP_KEY> as the bearer token. Keeping the real key out of this chat is intentional; I will paste it into the config file myself after you save it.

After saving, do not wait for the Settings > MCP panel to refresh. Tell me to replace <YOUR_MCP_KEY> with my real key, restart or reload Codex Desktop, wait up to 60 seconds for the MCP list to repaint, then test with: "List my DataUnmess datasets".

Claude, Cursor, Windsurf

Remote MCP clients. Copies the general remote-MCP installer prompt for clients that use JSON-style MCP config.

Preview prompt

I already have a DataUnmess MCP key. Install the DataUnmess MCP server into this AI client. Fetch https://www.dataunmess.ai/connect-mcp for the exact client-specific config snippet, then add it to this client's MCP config.

Keep the literal string <YOUR_MCP_KEY> as the bearer token. Keeping the real key out of this chat is intentional; I will paste it into the config file myself after you save it.

After saving, remind me to replace <YOUR_MCP_KEY> with my real key, restart or reload the AI client, and test with: "List my DataUnmess datasets".

Gemini or ChatGPT

When MCP is available. Use this same prompt in any AI client that supports remote MCP servers and config editing.

Preview prompt

I already have a DataUnmess MCP key. Install the DataUnmess MCP server into this AI client. Fetch https://www.dataunmess.ai/connect-mcp for the exact client-specific config snippet, then add it to this client's MCP config.

Keep the literal string <YOUR_MCP_KEY> as the bearer token. Keeping the real key out of this chat is intentional; I will paste it into the config file myself after you save it.

After saving, remind me to replace <YOUR_MCP_KEY> with my real key, restart or reload the AI client, and test with: "List my DataUnmess datasets".

Ask DataUnmess.ai MCP to start

Once the MCP server is connected, use one of these first actions. Start with company discovery, or go straight to a dashboard.

Start from your company site

Copy this after your MCP server is connected.

Preview prompt

Start my DataUnmess.ai MCP onboarding. Use my company site at https://example.com and docs at https://example.com/docs to understand the business, save useful company context with DataUnmess knowledge tools, then propose three dashboards we should create first.

Create a dashboard from a file

Copy this after your MCP server is connected.

Preview prompt

Use DataUnmess.ai MCP to create a dashboard for this file: C:\tmp\sales.csv. Upload it, inspect the columns, then build KPI cards, a revenue trend, and a breakdown by region.

Use existing workspace data

Copy this after your MCP server is connected.

Preview prompt

Use DataUnmess.ai MCP to list my datasets and connections, explain what is available, and suggest the best first dashboard to build.

What happens next

DataUnmess turns the first ask into next steps.

A good first action should either learn context or create a useful artifact. The MCP tools can save company knowledge, inspect datasets, upload files, build charts, and return links to dashboards the team can keep editing.

Learn company context

Create initial data catalog

Propose dashboards to build

Manual config and advanced setup

Use this only if you want to edit the client config yourself. Replace <YOUR_MCP_KEY> with the key from step 1, then restart or reload your AI client.

Codex Desktop

Config file:

macOS/Linux: ~/.codex/config.toml
Windows: %USERPROFILE%\.codex\config.toml

Add this TOML block:

[mcp_servers.dataunmess]
url = "https://mcp-dev.dataunmess.ai/mcp"
http_headers = { Authorization = "Bearer <YOUR_MCP_KEY>" }

Remote MCP JSON clients

Config file:

Claude Desktop: ~/Library/Application Support/Claude/claude_desktop_config.json
Cursor: ~/.cursor/mcp.json
Windsurf: ~/.codeium/windsurf/mcp_config.json

Add this entry:

{
  "mcpServers": {
    "dataunmess": {
      "url": "https://mcp-dev.dataunmess.ai/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_MCP_KEY>"
      }
    }
  }
}

Endpoint: https://mcp-dev.dataunmess.ai/mcp. Codex Desktop can take up to 60 seconds to repaint Settings > MCP after config changes.

query_data

Filter, group, and aggregate datasets (CSV, Excel, uploads). Returns result rows as JSON.

build_chart

Generate a chart spec from queried data. 22 chart types: bar, line, donut, scatter, treemap, funnel, radar, sankey, heatmap, waterfall, gantt, and more.

build_analysis

Create an analysis panel with text summary and KPI highlight cards showing metrics and trends.

query_connection

Execute a SELECT SQL query against a connected database (PostgreSQL, MySQL). Read-only — write queries are blocked.

query_github

Browse a connected GitHub repo: tree, file contents, languages, code search.

query_google_sheets

Read a connected Google Sheets / Drive source. Actions: list_spreadsheets, get_metadata, read_range, read_sheet_as_table. Read-only OAuth scopes.

build_flowchart

Build a flowchart with 8 node shapes (rect, diamond, ellipse, hexagon, parallelogram, cylinder, cloud, document), optional lucide icons per node, and labeled edges. Decision diamonds with multiple outgoing edges auto-fan into parallel lanes. Accepts structured nodes+edges JSON or a Mermaid `flowchart TD|LR` source block.

Four resources are also exposed: dash-ai://datasets, dash-ai://datasets/{id}, dash-ai://connections, and dash-ai://connections/{id}. Read dash-ai://connections first to discover available data sources before calling query_* tools.