
Danube AI
- Updated March 1, 2026
- danubeai/danube-mcp
Danube AI is a MCP server that provides unified AI tool discovery, execution, and integration across APIs and MCP servers via mcp.danubeai.com.
About
Danube AI MCP is a hosted Model Context Protocol server that acts as a unified layer for discovering, running, and wiring AI tools across REST APIs and other MCP endpoints. developers often hit integration fatigue when every service needs its own key, schema, and client code; Danube centralizes that behind one remote MCP connection authenticated with a Danube API key. Use it while building agent products, internal copilots, or automation stacks where the agent must pull from many backends without you cloning MCP configs for each vendor. The server.json specifies streamable-http remotes and a required secret header, so setup is register URL plus API key rather than cloning a monorepo of adapters. It is an integration platform MCP, not a substitute for security review or analytics dashboards.
- Remote streamable-HTTP MCP at mcp.danubeai.com/mcp
- Unified discovery, execution, and integration across APIs and MCP servers
- Requires danube-api-key header (issued at danubeai.com dashboard)
- Published version 1.0.1; GitHub: danubeai/danube-mcp
Danube AI by the numbers
- Data as of Jul 7, 2026 (Skillselion catalog sync)
claude mcp add --transport http danube https://mcp.danubeai.com/mcp --header "danube-api-key: YOUR_DANUBE_API_KEY"Add your badge
Show developers this MCP server is listed on Skillselion. Paste this into your README.
| Transport | HTTP |
|---|---|
| Auth | Required |
| Last updated | March 1, 2026 |
| Repository | danubeai/danube-mcp ↗ |
What it does
Discover and execute external APIs and other MCP tools from one Danube remote server so your agent does not maintain dozens of separate integrations.
Who is it for?
Agent developers who want one MCP remote to orchestrate multiple APIs and MCP tools during product development.
Skip if: Skip if you only need one fixed integration with no discovery layer, or teams without a Danube API key budget.
What you get
After adding the remote MCP URL and danube-api-key, your agent can discover and run integrated tools through Danube’s single protocol entry point.
- Single remote MCP registration for Danube tool catalog and execution
- Reduced per-integration MCP boilerplate in the agent project
By the numbers
- Server version 1.0.1 in published MCP schema
- Remote type: streamable-http at mcp.danubeai.com/mcp
- Required header: danube-api-key (secret)
README.md
Danube AI MCP Server
Danube is a unified platform for AI tool discovery, execution, and integration. This MCP server provides access to 30+ tools for service discovery, tool execution, workflows, skills, credentials, wallet management, and more.
Connection
URL: https://mcp.danubeai.com/mcp
Transport: Streamable HTTP
Authentication
Requires a Danube API key passed via the danube-api-key header. Get one at danubeai.com/dashboard/api-keys.
Configuration
Claude Desktop / Claude Code
{
"mcpServers": {
"danube": {
"type": "streamable-http",
"url": "https://mcp.danubeai.com/mcp",
"headers": {
"danube-api-key": "dk_your_api_key"
}
}
}
}
Available Tools
| Category | Tools |
|---|---|
| Discovery | list_services, search_tools, get_service_tools, get_recommendations |
| Execution | execute_tool, batch_execute_tools |
| Wallet | get_wallet_balance, get_spending_limits, update_spending_limits |
| Credentials | store_credential |
| Skills | search_skills, get_skill, create_skill, update_skill, delete_skill |
| Workflows | list_workflows, create_workflow, update_workflow, delete_workflow, execute_workflow, get_workflow_execution |
| Sites | search_sites, get_site_info |
| Agents | register_agent, get_agent_info, fund_agent_wallet |
| Ratings | submit_rating, get_my_rating, get_tool_ratings, report_tool |
SDKs
- Python: danubeai/danube-python |
pip install danube - TypeScript: danubeai/danube-ts |
npm install danube
Links
Recommended MCP Servers
How it compares
Meta-integration MCP hub, not a single-purpose database or analytics MCP.
FAQ
Who is Danube AI MCP for?
Developers shipping agent features who need centralized tool discovery and execution across APIs and MCP servers.
When should I use Danube AI MCP?
Use it during build when composing multi-tool agents and you want one remote MCP instead of many per-vendor servers.
How do I add Danube AI MCP to my agent?
Create an API key at https://danubeai.com/dashboard/api-keys, add remote URL https://mcp.danubeai.com/mcp with the danube-api-key header in your MCP client, per danubeai/danube-mcp on GitHub.