Visual query builder
Queries youcan read.
Build key conditions and filters as pills that know your key schema. Dynomatic compiles the optimal native expression underneath and lands the results in a table you can edit in place.
No expression syntax required. Pills know your key schema: real attribute names and typed values, not #placeholders and :bindings. Dynomatic compiles the optimal expression underneath, in the open.
Query when you can, Scan when you mean to. The builder reads your table's key schema and indexes, offers the access paths that exist, and never quietly downgrades a Query into a Scan. See what a read costs →
Every type, its own operators. Pick an attribute and the operator menu offers what DynamoDB can test on its type. Strings and binary take comparisons, BEGINS WITH, CONTAINS, ranges, size, and type tests. Numbers take comparisons, ranges, and presence. Sets and lists match by member with CONTAINS ANY and CONTAINS ALL, measure with SIZE, SIZE BETWEEN, and IS EMPTY, and never offer a whole-value equals. Binary values enter as Base64 or Hex.
Open any item as JSON or as a document. The Edit sheet sits beside the table with the whole item as JSON, and one toggle flips it to the exact DynamoDB shape, typed attribute values and all. The typed shape is also the only place to create sets (SS, NS, BS) and big ints, because plain JSON can't express them: a set is indistinguishable from a list, and numbers past 2⁵³ lose precision. Switch to Document for one row per attribute, keys first, each with its type and its value drawn the way the grid draws it.
Nothing writes until you say so. Edits stage into a pending set with a visible diff. Review every change, revert any of them, and commit when it is right, with prod labeled in red the whole way.
The builder, up close
PK and SK pills validated against your key schema. The partition key takes =, and BEGINS WITH is offered only on a string or binary sort key.
Stack filters on any attribute type with AND and OR, grouped visually, compiled to a single native expression. IS NULL, TYPE IS, SIZE, and set membership are a menu pick away. The attribute picker lists what the table is known to hold before your first read.
Primary and every GSI and LSI, listed with their key attributes so you pick a path, not a guess.
Open any item beside the table as JSON or as a document, and step to the previous or next row without closing it.
A strip under the builder estimates the RCU the next Run will read, before you press it. RCU, item count, and latency on every run, recorded in command history. A Limit set for one query replaces the table default for its Run and Load more.
Results to JSON, CSV, the clipboard, or straight to S3, one menu away.