CSV format

The first line names the columns. Every line after it is one capture. Resurf matches columns by their exact name, in any order, and ignores columns it doesn't know, so you only need the ones you use.

The only required column is type. Blank lines are skipped. In messages, the header is row 1, so your first capture is row 2.

Download the template CSV to start from a file that already imports.

Three captures: a link, a note whose text is in a Markdown file, and an image. The note and image files sit in folders next to the CSV, and their paths are relative to the CSV's folder.

my-import/
  captures.csv
  notes/launch-checklist.md
  assets/board.png

type,title,source_url,user_note,tags,spaces,created_at,note_path,asset_path
link,How to price a product,https://example.com/pricing,Read before the launch,pricing;research,Research,2024-03-01T09:30:00Z,,
note,Launch checklist,,,launch,Projects,2024-03-02T18:00:00+01:00,notes/launch-checklist.md,
image,Mood board,https://example.com/board,,design;inspiration,Design,2024-03-03T12:00:00Z,,assets/board.png

Empty cells are fine. The link gets a Note, two tags and a space. The note's date uses a time zone offset; the others use Z for UTC.

Write the type exactly as shown: lower case, except voiceNote.

Anything else, including Link or URL, makes the row invalid.

• The file must be UTF-8. A byte order mark at the start is fine. In Excel, save as CSV UTF-8 (Comma delimited).
• Commas separate columns. A file that uses semicolons or tabs between columns won't import.
• Wrap a cell in double quotes when it contains a comma, a quote or a line break, and double any quote inside it: "She said ""yes"", then left".
• Column names are lower case and exact: Type or Source URL won't match.
• Dates need a time and a time zone: 2024-03-01T09:30:00Z or 2024-03-01T09:30:00+01:00. A date alone, like 2024-03-01, makes the row invalid.
• A note's text can't go in a cell. It comes only from a file (note_path) or a web address (note_url). For a short line of your own, use user_note.
• A note or quote row without note_path or note_url imports with an empty body. An image, PDF, video, audio or file row without asset_path or asset_url imports with no file.
• Without a license, Resurf stops an import that would go past 50 captures before it adds anything. On the free plan you can have 4 spaces; space names past that are left out without a message.

A Resurf export writes these 42 columns, in this order.

capture_id,external_id,schema_version,type,title,description,user_note,source_kind,source_url,domain,source_app,source_file_name,tags,spaces,space_ids,primary_space,primary_space_id,created_at,updated_at,device_name,pinned_at,later_at,deleted_at,note_path,note_url,asset_path,asset_url,mime_type,original_filename,source_json,organization_json,state_json,content_json,files_json,comments_json,ai_suggested_title,ai_suggested_tags,ai_suggested_spaces,ai_tldr,ai_source_fingerprint,ai_processor_version,ai_generated_at

The columns ending in _json and starting with ai_ carry what only Resurf knows: comments, files, AI summaries. Leave them out of files you build yourself. When one is present, Resurf reads it, so a hand-edited one that isn't valid JSON makes the row invalid.

The preview shows up to three problems. The toast after the import counts what failed.