MCP

Compass exposes an MCP (Model Context Protocol) server at /mcp. Any MCP-compatible AI tool can connect and query the knowledge graph using structured tools.

Setup

Claude Code

Add to your project's .mcp.json:

{
  "mcpServers": {
    "compass": {
      "type": "sse",
      "url": "http://localhost:8080/mcp"
    }
  }
}

Cursor

Add to Cursor's MCP settings (Settings > MCP Servers):

{
  "compass": {
    "type": "sse",
    "url": "http://localhost:8080/mcp"
  }
}

Windsurf

Add to Windsurf's MCP configuration:

{
  "mcpServers": {
    "compass": {
      "serverUrl": "http://localhost:8080/mcp"
    }
  }
}

Tools

`search_entities`

Search the knowledge graph with keyword, semantic, or hybrid mode.

Parameter	Required	Description
`text`	Yes	Search query
`types`	No	Comma-separated entity types (e.g., `table,topic`)
`source`	No	Filter by source system
`mode`	No	`keyword`, `semantic`, or `hybrid` (default: `keyword`)
`size`	No	Max results (default: 10)

`get_context`

Get full context about an entity: the entity itself, its relationships, and related entities.

Parameter	Required	Description
`urn`	Yes	Entity URN
`depth`	No	Traversal depth, 1-5 (default: 2)

`impact`

Analyze downstream blast radius — what entities are affected if this entity changes.

Parameter	Required	Description
`urn`	Yes	Entity URN
`depth`	No	Downstream traversal depth (default: 3)

`get_documents`

Get documents (runbooks, annotations, decisions) attached to an entity.

Parameter	Required	Description
`urn`	Yes	Entity URN

Usage Patterns

AI agents typically compose these tools together:

Discover — search_entities to find relevant entities
Understand — get_context to see relationships and neighbors
Assess risk — impact to evaluate downstream effects
Get knowledge — get_documents for runbooks and decisions

The agent controls what context it assembles. Compass provides focused, composable tools rather than monolithic responses.