01 — Overview
Document databases are quick for engineers and hard for everyone else. A support agent shouldn’t have to read raw JSON to find out why an order is stuck. Describe the view you need, and the Bench samples the collection, works out which fields matter and lays them out as tables, detail panels and edit forms.
Nested arrays become sub-tables, and references between collections become links. For reporting, the Bench writes aggregation pipelines you can read and edit in the Code tab. Each update records the fields it changed and their old values in the Logbook, so a mistaken edit can be traced and reversed. Row-level rules work on document fields too, so a regional team sees only its own customers. Connect Atlas or a self-managed cluster, and keep writes on a staging cluster until a reviewer approves the release.
02 — What you can do
Schema from samples
The Bench reads a sample of documents to infer field types, even in uneven collections.
Nested data, laid flat
Arrays and subdocuments render as sub-tables and grouped fields you can edit in place.
Field-level history
Every update stores changed fields and old values, so you can trace and revert a bad edit.
03 — FAQ
Before you connect MongoDB
Does Chamfer support MongoDB Atlas?
What if our documents don't share a schema?
Can we write aggregation pipelines ourselves?
How do we stop edits to the wrong fields?
04 — Pairs well with
