Available Tools

The Indigo MCP server exposes 4 tools. Claude (or any MCP client) calls these tools automatically when you ask questions about your data.

Tool overview

query_collection

Query any collection with filters, projections, sorting, and pagination. This is the primary tool for most data access.

Parameters

Parameter	Type	Required	Default	Description
collection	String	Yes	—	`signals`, `meetings`, `meetingdata`, or `taxonomy`
filter	Object	No	`{}`	MongoDB filter conditions
projection	Object	No	all fields	Fields to include (1) or exclude (0)
sort	Object	No	—	Sort order: 1 ascending, -1 descending
limit	Number	No	20	Max documents to return (1–100)

Example

Find pending action items, newest first:

{
  collection: "signals",
  filter: {
    insightType: "action",
    "data.status": "pending"
  },
  sort: { createdAt: -1 },
  limit: 10
}

aggregate_collection

Run a MongoDB aggregation pipeline for complex analytics — grouping, counting, cross-collection lookups, and computed fields. Tenancy filters are injected automatically.

Parameters

Parameter	Type	Required	Description
collection	String	Yes	`signals`, `meetings`, `meetingdata`, or `taxonomy`
pipeline	Object[]	Yes	Array of aggregation stages

Supported stages

Stage	Description
`$match`	Filter documents
`$group`	Group by field(s) and compute aggregates
`$sort`	Sort results
`$project`	Reshape documents, include/exclude fields
`$lookup`	Join with another collection
`$unwind`	Flatten arrays into separate documents
`$limit`	Cap result count
`$skip`	Skip N documents
`$count`	Count matching documents
`$addFields`	Add computed fields

Example

Count signals by type from the last 30 days:

{
  collection: "signals",
  pipeline: [
    {
      $match: {
        createdAt: {
          $gte: { $date: "2025-01-01T00:00:00Z" }
        }
      }
    },
    {
      $group: {
        _id: "$insightType",
        count: { $sum: 1 }
      }
    },
    { $sort: { count: -1 } }
  ]
}

list_collections

Returns metadata about all available collections, including document counts visible to your account.

Parameters

None.

Response

[
  { name: "signals", description: "...", count: 342 },
  { name: "meetings", description: "...", count: 87 },
  { name: "meetingdata", description: "...", count: 64 },
  { name: "taxonomy", description: "...", count: 156 }
]

collection_info

Returns detailed schema documentation for a specific collection — field types, relationships, and example queries.

Parameters

Parameter	Type	Required	Description
collection	String	Yes	`signals`, `meetings`, `meetingdata`, or `taxonomy`

Response

Returns a markdown-formatted schema document including field definitions, type-specific structures, access control patterns, and example queries. This is what Claude uses to understand how to query your data.

When to use which tool

Scenario	Tool
Find specific records by criteria	`query_collection`
Count, group, or aggregate data	`aggregate_collection`
Join data across collections	`aggregate_collection` with `$lookup`
Simple list with filters and sorting	`query_collection`
Check what data is available	`list_collections`
Understand a collection’s schema	`collection_info`