ScoreSifter Kairovo ↩
v1.0 — free
KA-101 / ScoreSifter · Guide

Using ScoreSifter

Everything you need to get from a messy folder to a clean, forScore-ready library.

§ 01

Getting started

Download ScoreSifter for your platform and open it. No account, no licence key, no internet connection required.

On first launch, you'll see the onboarding screen. It walks you through pointing the app at your score folder — a folder of PDFs on your computer. This can be any folder; ScoreSifter doesn't copy or move your files, it just reads and renames them in place.

System requirements

  • macOS — 11.0 (Big Sur) or later, Apple Silicon or Intel
  • Windows — Windows 10 or 11, x64
  • Linux — AppImage, tested on Ubuntu 22.04 and Fedora 38

Opening a folder

  1. Click Open Folder or drag a folder onto the app window
  2. ScoreSifter scans every PDF in the folder (and subfolders)
  3. Each file is parsed and matched against the catalogue
  4. Results appear in three tabs: To Review, Done, and All
Good to know: ScoreSifter never changes a file until you explicitly commit a rename. Everything up to that point is preview only.

The three tabs

  • To Review — files that need your attention. Low-confidence matches, unmatched files, or anything that looks wrong.
  • Done — files you've reviewed and approved. These are queued for renaming when you click Rename Selected.
  • All — every file ScoreSifter found, regardless of status.

Row controls

Each file row has three icon buttons on the right, visible on hover:

Icon What it does
🔍 Search Opens a manual lookup for that file. Use this when auto-lookup didn't find a match, or when you want to search for a different result. Results come from the local catalogue first, then MusicBrainz if nothing is found.
👁 Eye Shows the original filename as it was before ScoreSifter parsed it. Useful when you want to double-check what the file was called before any cleanup.
↺ Reset Clears the proposed rename for that file and restores it to its auto-detected state. Use this if you've edited a row and want to start over.

§ 02

How the catalogue works

ScoreSifter ships with KairovoData — a hand-curated database of around 43,000 works covering art song, opera arias, lieder and musical theatre. The database is embedded in the app and works entirely offline.

Confidence levels

Each file gets a confidence rating based on how cleanly ScoreSifter could match it:

  • High (green) — strong match from the catalogue. Title and composer extracted cleanly, verified against KairovoData.
  • Medium (amber) — probable match, but something was ambiguous. Worth a quick check.
  • Low (grey) — ScoreSifter made a guess. Review these manually before committing.

Confidence reflects the quality of the filename parse — not whether the proposed rename looks right to you. Always review low-confidence files before committing.

MusicBrainz fallback

When a file doesn't match anything in the local catalogue, ScoreSifter can optionally query MusicBrainz — a public, community-maintained music database. This sends only the title and composer name, nothing else.

MusicBrainz lookups are marked separately so you can tell which results came from the local catalogue and which came from online. You still review and approve each one before it's used.

Manual entry

If nothing matches — catalogue or MusicBrainz — you fill in the fields yourself. Type the title, composer, key, and work directly into the file row. ScoreSifter renames from your input, the same as any other file.

§ 03

Templates

Templates control how the renamed filename is constructed. ScoreSifter comes with five presets and lets you write your own.

Available tokens

{title}
The work title — e.g. "Die Forelle"
Required
{composer}
Composer surname — e.g. "Schubert"
Optional
{key}
Key signature — e.g. "D", "A♭", "c"
Strongly recommended
{work}
Parent opera or show — e.g. "Rigoletto"
Optional — use with [? … ?]

Optional segments

Wrap any part of a template in [? … ?] to make it conditional — it only appears if the token inside has a value.

For example: {title} – {composer}[? ({work})?] – [{key}]

If the piece has a parent work, you get: Dido's Lament – Purcell (Dido and Aeneas) – [g].pdf

If it doesn't, you get: Die Forelle – Schubert – [D].pdf

Why include the key? When the same piece exists in multiple keys in your library, the key in the filename is the only way to tell them apart at a glance — and it's what ScoreSifter uses to suggest the right file when building a Recital Atlas program.

§ 04

forScore integration

ScoreSifter doesn't connect to forScore directly — it works on your PDF files before they ever reach the iPad. By the time you import your scores into forScore, all the metadata is already in the files, waiting to be fetched.

What gets written into each PDF

After a rename, ScoreSifter writes metadata into each PDF using the standard fields forScore reads:

PDF field forScore reads as Example
Title Title La donna è mobile
Author Composer Giuseppe Verdi
Keywords Key signature + tags keysf:2, keymi:0, Rigoletto

Key signatures use the MIDI standard — keysf for the number of sharps or flats, keymi for major (0) or minor (1). This is the same format defined in the forScore PDF metadata spec.

Step by step: getting metadata into forScore

  1. 01
    Rename your files with ScoreSifter Scan your folder, review the proposed renames, and commit. ScoreSifter renames the files and writes the metadata in one step.
  2. 02
    Import the files into forScore Add your renamed PDFs to forScore using any method you prefer — AirDrop, Files app, iCloud Drive, or the forScore file import. The files now contain embedded metadata.
  3. 03
    Fetch the metadata in forScore Open any score in forScore. Tap the menu button (the three dots or the title bar, depending on your forScore version) and choose Fetch. forScore reads the embedded PDF metadata and populates Title, Composer, and Key immediately.
  4. 04
    Or, turn on automatic fetching To have forScore fetch metadata automatically for every new file you add, go to Settings → Score Library → Automatic Metadata and enable Automatic fetching for new files. Once this is on, you never need to tap Fetch manually — import and it's done.
If Fetch doesn't seem to work: Check that the file was actually renamed and committed in ScoreSifter (not just previewed). Files in the "To Review" tab haven't been renamed yet — move them to Done and commit first.

What forScore does with the work field

forScore doesn't have a dedicated "work" or "opera" field in its metadata panel — it maps the PDF Subject field to Genre. ScoreSifter writes the parent work (e.g. "Rigoletto") as a keyword tag instead, which appears in forScore's tags system. You can use these tags to group all arias from the same opera.

Compatible readers

The metadata ScoreSifter writes uses standard PDF fields — not forScore-specific formats. It also works with MuseScore, MobileSheets, and other music PDF readers that parse standard PDF metadata.

§ 05

Undo

Every rename run ScoreSifter performs is logged with the full old → new mapping. You can reverse a batch — or individual files within it — as long as the app is open.

How to undo

  1. Go to the All tab
  2. Select the files you want to reverse, or select all files from a run
  3. Click Undo Renames

ScoreSifter reverses the rename and removes the metadata it wrote from each PDF. The files return to their original names.

Once the app is closed, the undo log is cleared. If you need to reverse renames after closing ScoreSifter, you'll need to do it manually or restore from a backup. This is why we recommend keeping a backup of your score folder before running a large batch for the first time.

Partial undo

You don't have to undo an entire run. Select individual files in the All tab and undo just those — the rest of the batch stays as renamed.