Schema beside your records
Inspect the embedded writer schema, named types and logical-type annotations alongside your data.
Open an Avro Object Container File locally. Search its records, inspect blocks and codecs, and explore its embedded writer schema.
Inspect records, embedded schemas, blocks and codecs. Record tables also support CSV. Large files export the displayed page as JSON. JSON/CSV exports are limited to 10 MB. File contents are decoded locally; normal page requests reach the hosting provider.
Inspect the embedded writer schema, named types and logical-type annotations alongside your data.
Search and filter records, then export JSON or CSV using the existing Avro workspace.
Uncompressed, deflate and Snappy blocks are supported. Sync markers and Snappy checksums are checked; unsupported files show an error.
Temporary handoffs use local browser storage and expire after 30 minutes. Refresh can reopen the original file while its handoff remains available; download edits before reloading. Back to website preserves the live workspace for explicit Resume.
Inspect records alongside their embedded writer schema, named types and logical-type annotations. The workspace also shows container blocks and codecs and provides record search, per-field filters and sorting.
Choose an Avro Object Container File (.avro) containing its embedded writer schema. Uncompressed, deflate and Snappy blocks are supported. Sync markers and Snappy checksums are checked. Unsupported codecs or invalid containers show an error; standalone binary records without a container are not supported.
Local editing decodes files in your browser. Optional assistance sends only your goal, an explicitly included schema and relevant requested record count to OpenRouter and the configured model provider; OCF record values are not included automatically. Original-input handoffs use local browser storage and expire after 30 minutes. Normal page and sample requests reach the hosting provider.
Yes. The decoded record table exports all matching applied records across pages as JSON or CSV, limited to 10 MB; unapplied drafts are excluded. Download a copy includes all applied records, including filtered-out records. Large-file page exports contain only the displayed saved-source records and exclude staged edits. Copies leave source edits unsaved. You can also export the embedded writer schema as an .avsc file below 1 MB.