Writing, import and voice
Bring the documents you already have, write new ones in the app, or upload a recorded take. Sources are hashed and versioned and never changed, so every citation points at text that really exists.
Import Markdown, text and Word
Drop .md, .txt, .fountain, .yarn, .ink, .csv, .tsv, .json or .docx files anywhere in the Library. Word files keep headings, lists, tables and footnotes; tracked changes are read as accepted, and Word comments become notes pinned to the words they were on. The original Word file is kept for download.
- Same filename is a new version, same text is skipped
- Folders, tags and kinds to keep a large Library in order
- A version diff shows which canon facts cite changed passages
Read, write and add facts by hand
Read any document with find, outline and a version picker. Write or paste new documents in the app; drafts autosave privately. Select words to add a fact, an entity or a note, and run a continuity check of a draft against canon.
- Add a fact from one sentence; who, where and when are filled in for you to check
- Notes sit in the margin until they are done
- When a new version changes a cited passage, the fact is flagged for re-check
Recordings as sources
Upload a voice-actor take, a table read or an interview. The transcript becomes a document with timed lines, and its citations play the moment they quote. Link the script, and CanonTracker compares what was said with what was written and flags lines that contradict canon.
- MP3, WAV, M4A, OGG, WebM and FLAC, up to 30 minutes
- Dictate into the editor and the Ask box
- Your project’s names help the speech recognition
Start with a finished story
Create a studio and load The Ashbell Pass or Broken Tide: real documents, AI candidates to review and planted contradictions to find. Or run CanonTracker on your own server; it is open source.