Diff first, then merge
If you are not sure B will drop a field, compare first. Merge is the action; Diff is the decision.
JSON Tools for AI Developers · Runs locally
Natural language generation · Schema · Zod · OpenAPI · Function Calling · MCP · 30+ tools
One file is defaults, another is the environment overlay. You only see the effective object after merge.
The left is base A, the middle is overlay B. Same-name keys come from B. Keys only in A are kept.
Arrays are replaced as a whole — they are not concatenated by element. Use Diff to see which keys changed.
Both documents are merged locally. Nothing is uploaded.
If you are not sure B will drop a field, compare first. Merge is the action; Diff is the decision.
Concat vs replace has no single business meaning. A user list and a feature-flag array are different, so the default is replace.
The whole array is replaced by default. Concatenate by hand, or split the merge into steps.
Objects nest downward. Exact depth follows the current implementation — scan nested results after merge.
Diff first so you see the gap, then decide which side wins.
No.
Same-name keys are written from B, including null. Delete the key from B if you don't want that.
Two at a time. Merge A+B, then treat the result as the new A and merge C.
Write order follows the implementation. Sort keys after merge if you need a stable Diff.
B wins. If A is an object and B is a string, the result is a string.