Skip to main content

Tools Reference

Complete reference for all platform tools available via MCP

The Lovelace MCP server exposes the following tools. Each tool accepts a JSON input and returns structured results.

Workspace Tools

lovelace_list_workspaces

List all workspaces accessible to the authenticated user.

Input: None required

Output:

json
{
  "workspaces": [
    {
      "id": "ws_abc123",
      "name": "My Project",
      "description": "Main development workspace",
      "createdAt": "2025-01-15T10:00:00Z",
      "updatedAt": "2025-02-01T14:30:00Z"
    }
  ]
}

lovelace_get_workspace

Get detailed information about a specific workspace.

Input:

ParameterTypeRequiredDescription
workspaceIdstringYesThe workspace ID

Example:

json
{
  "workspaceId": "ws_abc123"
}

Output:

json
{
  "id": "ws_abc123",
  "name": "My Project",
  "description": "Main development workspace",
  "members": 3,
  "agentCount": 2,
  "createdAt": "2025-01-15T10:00:00Z",
  "updatedAt": "2025-02-01T14:30:00Z"
}

Agent Tools

lovelace_list_agents

List available agents and their current status.

Input: None required

Output:

json
{
  "agents": [
    {
      "id": "agent_xyz789",
      "name": "code-reviewer",
      "status": "idle",
      "description": "Reviews code for quality and security issues"
    }
  ]
}

lovelace_spawn_agent

Spawn an agent to execute a task.

Input:

ParameterTypeRequiredDescription
agentIdstringYesThe agent to spawn
taskstringYesTask description for the agent
workspaceIdstringNoWorkspace context for the task
configobjectNoAdditional agent configuration

Example:

json
{
  "agentId": "code-reviewer",
  "task": "Review the authentication module for security vulnerabilities",
  "workspaceId": "ws_abc123"
}

Output:

json
{
  "executionId": "exec_def456",
  "agentId": "code-reviewer",
  "status": "running",
  "startedAt": "2025-02-17T12:00:00Z"
}

lovelace_get_agent_status

Check the execution status of a spawned agent.

Input:

ParameterTypeRequiredDescription
executionIdstringYesThe execution ID returned by lovelace_spawn_agent

Example:

json
{
  "executionId": "exec_def456"
}

Output:

json
{
  "executionId": "exec_def456",
  "agentId": "code-reviewer",
  "status": "completed",
  "startedAt": "2025-02-17T12:00:00Z",
  "completedAt": "2025-02-17T12:05:30Z",
  "progress": 100
}

Status values: queued, running, completed, failed, cancelled


lovelace_get_agent_result

Retrieve the output and artifacts from a completed agent execution.

Input:

ParameterTypeRequiredDescription
executionIdstringYesThe execution ID

Example:

json
{
  "executionId": "exec_def456"
}

Output:

json
{
  "executionId": "exec_def456",
  "status": "completed",
  "output": "Found 3 potential security issues in the authentication module...",
  "artifacts": [
    {
      "name": "security-report.md",
      "type": "text/markdown",
      "size": 4096
    }
  ]
}

Knowledge Tools

lovelace_search_knowledge

Search the knowledge base for relevant documents.

Input:

ParameterTypeRequiredDescription
querystringYesSearch query
limitnumberNoMaximum results (default: 10)
workspaceIdstringNoLimit search to a specific workspace

Example:

json
{
  "query": "authentication best practices",
  "limit": 5
}

Output:

json
{
  "results": [
    {
      "id": "doc_ghi789",
      "title": "Authentication Architecture",
      "snippet": "Our authentication system uses OAuth 2.1 with PKCE...",
      "score": 0.95,
      "updatedAt": "2025-01-20T09:00:00Z"
    }
  ],
  "total": 12
}

lovelace_store_knowledge

Store a document in the knowledge base.

Input:

ParameterTypeRequiredDescription
titlestringYesDocument title
contentstringYesDocument content (Markdown supported)
workspaceIdstringNoWorkspace to store the document in
tagsstring[]NoTags for categorization

Example:

json
{
  "title": "API Rate Limiting Policy",
  "content": "# Rate Limiting\n\nAll API endpoints enforce rate limits...",
  "tags": ["api", "policy", "security"]
}

Output:

json
{
  "id": "doc_jkl012",
  "title": "API Rate Limiting Policy",
  "createdAt": "2025-02-17T12:10:00Z"
}

Error Handling

All tools return errors in the standard MCP error format:

json
{
  "content": [
    {
      "type": "text",
      "text": "Error: Workspace not found"
    }
  ],
  "isError": true
}

Common error conditions:

ErrorCause
Workspace not foundThe specified workspace ID does not exist or is not accessible
Agent not foundThe specified agent ID does not exist
Execution not foundThe specified execution ID does not exist
Authorization requiredMissing or invalid authentication token
Insufficient permissionsYour token lacks the required scope