Free

Open related documents from your query results

Run one scan, and Monghoul finds the fields that point at other collections and checks them against your data. After it, an id in a result opens the document it names, and the editor offers the right $lookup first.

Download Monghoul

An order’s customerId opened as the customer document beside the result, and that customer’s referredBy opened as a second customer document — Fig 01 Results · An order’s customer, and the customer who referred them

The whole database on one screen

Each collection is a box of its fields and types, and a line joins each field to the collection it points at: solid when the scan verified it, dashed when it is only likely. Select a collection to list its links both ways.

The Database Map of a shop database: each collection as a box of its fields and types, lines between the fields that reference each other, and orders selected, with its three references and the two collections that reference it — Fig 02 Database Map · Every collection a box of its fields, and one selected lists its links both ways

Every link shows its evidence

Select a line to see why the scan believes it: how many sampled ids were found, whether the names and types agree, and whether the field has an index. You confirm the link or dismiss it, and its $lookup stage is ready to copy.

The evidence for orders.customerId to customers._id: verified, 31 of 32 sampled ObjectIds found, no other collection holds them, the name and types agree, used by a $lookup in two saved queries, indexed, and the $lookup stage to copy — Fig 03 Database Map · Why the scan believes a link, and the $lookup that follows it

Problems found on the way

References with no index, mixed types, ObjectIds stored as strings, and values that match nothing. Each finding ends in an action the app already has, such as Create index.

Findings from a scan: a reference with no index, a field with mixed types, ObjectIds stored as strings, and a reference with one unmatched value, each with its action — Fig 04 Database Map · What the scan found wrong, each with the action that fixes it

Also in this part

In the editor: Inside $lookup, the collections this one references come first, and localField and foreignField put the relationship’s two fields first. A snippet writes the whole stage.
For an assistant: On Pro, an AI assistant reads the same map, and find_join_path gives it the $lookup pipeline between two collections.
Verified, likely, or yours: Verified needs 20 values looked up, 90% of them found, and a name, a DBRef or a $lookup as well. You confirm or dismiss the rest.
Limits: Up to 1,000 documents per collection (500 on Free), one command at a time, each with a 3-second limit. The list of what is new since the last scan needs Pro.

Try it against your own data

38 MB installer. Free forever, and a 14-day Pro trial with no card.

Download Monghoul