Tracearr

Tracearr adapter notes for importing watch history into CW.

Tracearr is a read-only history importer. It reads play history from Tracearr's public API and normalizes it for CW. Tracearr can monitor Plex, Jellyfin, and Emby.

Experimental: Use Interactive Sync and check the results. One-way syncing is strongly recommended.

Connect Tracearr first. Use: Tracearr.

What it supports

• Direction: source-only
• History import for movies and episodes
• Rewatches: each play can remain a separate watch
• Writes: not supported

It does not support watchlists, ratings, progress, playlists, music, live TV, or photos. It cannot sync back to Tracearr or its Plex, Jellyfin, or Emby servers.

How indexing works

1. Retrieves pages from /history with 100 plays per page. It stops at the next-cursor boundary or safety limit.
2. Keeps only media_type values of movie or episode.
3. Keeps only plays Tracearr marks as watched. Tracearr uses its own thresholds, which default to 85%.
4. Uses the available TMDb, IMDb, and TVDb IDs for movies.
5. For episodes, retrieves show IDs from /media/{show_media_id} once per show and caches them.
6. For unlinked plays, borrows show IDs from a play with the same show title and server. Ambiguous titles are skipped.
7. Uses external IDs as primary keys. The source-server item key (plex, jellyfin, or emby) is the fallback.

Without rewatches, history is a de-duplicated snapshot. The newest play wins for each canonical key. With rewatches enabled on the pair, every play remains a separate entry.

Output shape

Movies and episodes include an ids object with available IMDb, TMDb, and TVDb IDs. They also include the source-server item key. Episodes use the show's IDs, never the episode's own IDs.

{
  "type": "episode",
  "ids": {
    "tmdb": "103516",
    "imdb": "tt12327578",
    "tvdb": "382389",
    "plex": "43892"
  },
  "show_ids": {
    "tmdb": "103516",
    "imdb": "tt12327578",
    "tvdb": "382389"
  },
  "series_title": "Star Trek: Strange New Worlds",
  "season": 4,
  "episode": 9,
  "watched_at": "2026-09-19T08:57:43Z"
}

Limits

• Plays unlinked to a library item can lack IDs. CW skips them and reports skipped_no_ids.
• Tracearr allows about 240 API requests per minute per key. Large first syncs can take several minutes.

Recommended workflows

1

Import into CW

Run a pair: Tracearr → CW with History enabled. Then use CW as the source for downstream services.

2

Import into an Editor dataset

Use an Editor dataset to clean history, fix matches, and correct metadata before syncing elsewhere.

Endpoints used

All requests use Authorization: Bearer trr_pub_....

• GET /api/v2/public/history — history pages with cursor paging. Supports optional user_id and server_id.
• GET /api/v2/public/media/{show_media_id} — show IDs for episodes.
• GET /api/v2/public/users — connection check and user picker.