fileviewer.dev opens a data file and shows you what is actually in it. Drop in a 50 MB JSON export, a YAML config, an XML feed, a CSV extract or a Markdown document, and read it as formatted code, as a collapsible tree, as a sortable table, as a rendered preview, or as a graph of how its parts reference one another. Every one of the 9 views runs on your own machine.
There is no upload step and no account, because there is no server to upload to. The parser, the validator and the converters are all compiled into the page you are reading, so a file you open is read by your browser and by nothing else. You can disconnect from the network and keep working — the app installs and runs offline. This is the reason to use it for a production database dump, an API response with live credentials in it, or a customer export you are not permitted to paste into a website. The privacy page explains exactly what is stored and what is sent, and the claim is enforced by tests rather than asserted in marketing.
The format is detected from the content rather than the file extension, so a
.txt holding JSON is still parsed as JSON, and a file with the wrong
extension is not silently treated as plain text.
20 conversion pairs are supported, and each one reports what it had to change. That matters more than it sounds: CSV has no nesting, XML has no arrays, and YAML carries comments that JSON cannot represent. Rather than dropping those quietly, the converter tells you which structures were flattened, renamed or lost, so you find out before the output reaches something that depends on it.
JSON can be checked against a JSON Schema from draft-06 through 2020-12, with errors reported against the line that caused them instead of a byte offset. JSONPath queries run over the parsed document with a visual builder for the syntax nobody remembers, and a structural diff compares two files by meaning, so reindenting or reordering keys does not show up as a change. Validate JSON, run a JSONPath query or compare two documents.
Close the tab and come back next week and the documents you had open are still there, along with your theme, your saved schemas and your saved queries. They are kept in your browser's own storage rather than on an account somewhere, which is why this works without a login and why nobody else can read them. Large files are parsed on a background thread so the interface keeps responding while a 50 MB document is being read, and the whole app installs to your machine and keeps working with no connection at all.
Loading the workspace. If this message stays, JavaScript is blocked — the file reading happens in your browser, so the app needs it to run.