In the
podcast_ratings category, which is opt-in — it is callable by name on any connection, but only advertised on tools/list when selected via ?include=podcast_ratings. See Tool sets & discovery.platform_slug, min_stars, or a since/until window; the summary always reflects the full corpus. For just the headline numbers on several podcasts at once, particle_podcast_resolve with include: ["ratings_summary"] is cheaper.
Inputs
Output
A markdown document with## Ratings for <podcast_slug> and an **Overall:** row (X.XX stars across N ratings), then one BulletKV per (platform, locale) aggregate (platform/locale: X.XX stars across N ratings). A ### Listener sentiment section carries the narrative paragraph when present. A ### Recent reviews section lists each review as N★ Title — Author (YYYY-MM-DD) with the review body indented underneath. When more pages exist, a horizontal rule and a **Cursor:** <value> line are appended.
Sample (podcast_slug="all-in", limit=2):
Example
Related
- REST equivalent:
GET /v1/podcasts/{id}/ratings. - For the headline numbers inline on resolve, use
particle_podcast_resolvewithinclude: ["ratings_summary"].