Quick Start
Connect to Lovelace and complete your first MCP workflow
This guide walks through the public Lovelace MCP gateway from authentication to a full agent workflow.
Prerequisites
- A Lovelace account
- An MCP-compatible client such as Claude Desktop, Claude Code, Cursor, or Windsurf
- A client that supports MCP Streamable HTTP and the MCP HTTP authorization flow, or a client where you can manually complete OAuth configuration
Step 1: Configure Your Client
Use the public gateway endpoint:
- URL:
https://mcp.uselovelace.com/mcp - Transport: Streamable HTTP
- Authorization: Let the client follow the gateway’s
401challenge and Protected Resource Metadata discovery flow
Claude Desktop
{
"mcpServers": {
"lovelace": {
"type": "streamable-http",
"url": "https://mcp.uselovelace.com/mcp"
}
}
}
Claude Code
claude mcp add lovelace --transport streamable-http \
--url https://mcp.uselovelace.com/mcp
For other clients, see Client Setup.
Step 2: Let the Client Authorize
On first connection, the gateway challenges the client and points it to:
https://mcp.uselovelace.com/.well-known/oauth-protected-resource
From there, the client discovers Lovelace Accounts as the authorization server, completes OAuth 2.1 authorization, and starts sending bearer tokens on MCP requests.
If your client needs manual OAuth setup rather than automatic discovery, use the metadata discovered from the protected resource and configure the client to request the MCP endpoint as the resource/audience.
Step 3: Verify Access
Ask your client:
List my Lovelace workspaces
That should invoke lovelace_list_workspaces and return the workspaces you can access.
Step 4: Launch an Agent
Pick a workspace ID from the previous step, then ask your client to spawn an agent in that workspace:
Spawn a Lovelace agent in workspace ws_abc123 with agent type general to review our API performance bottlenecks
The underlying MCP tool is:
{
"workspaceId": "ws_abc123",
"agentType": "general",
"task": "Review our API performance bottlenecks"
}
The gateway returns an agentId and initial status.
Step 5: Check Status and Results
Use the returned agentId to monitor progress:
Check the status of Lovelace agent agent_123
Then fetch the final result:
Get the result for Lovelace agent agent_123
The agent workflow uses:
lovelace_get_agent_statuswithagentIdlovelace_get_agent_resultwithagentId
Step 6: Read Matching Resources
You can also read the corresponding MCP resources directly:
lovelace://workspaceslovelace://workspaces/ws_abc123lovelace://agents/agent_123/statuslovelace://agents/agent_123/output
For knowledge documents discovered through search:
lovelace://knowledge/doc_456
What’s Next
- Tools Reference — Exact tool contracts and scopes
- Resources Reference — Exact resource URIs and payloads
- Authentication — Protected-resource discovery and OAuth guidance
- Local Server — Local MCP server for Lovelace CLI/daemon workflows