particle_podcast_get_episode

Bundled overview of one episode — speakers, top entities, segment/clip counts, and optional transcript, segments, entities, clips, and topics.

Return a bundled overview of one podcast episode: title, podcast, speakers (with entity slugs), top mentioned entities, and segment/clip counts.

By default the response is lean — counts plus the top mentioned entities. Request optional sections via include: segments for the structural outline with timestamps and transcript line ranges, entities for the complete entity list (instead of the top 20), clips for engagement-ranked highlight clips, topics for topic classifications with slugs, related for the five episodes from other shows most related to this one (for the full ranked list with the reasons behind each match use particle_podcast_list_related_episodes), or transcript for the dialogue transcript (narrow it by speaker or time range via transcript_speaker / transcript_start / transcript_end — full transcripts are large).

This is also where a known episode’s full clip list lives — include: ["clips"]. There is no separate clip-listing tool; relevant clips for a search arrive inline on particle_podcast_search_transcripts matches.

For “every line about X in this episode” use particle_podcast_find_mentions with episode_slug instead — that returns the dialogue around each mention with is_mention flags.

Inputs

Field	Type	Required	Default	Description
episode_slug	string	yes	—	Episode slug or canonical ID.
include	array of enums	no	[]	Optional response sections: transcript (bounded dialogue transcript — large for long episodes; narrow with the transcript_* sub-params), segments (structural outline with timestamps and transcript line ranges), entities (complete mentioned-entity list instead of the top 20), clips (engagement-ranked highlight clips), topics (topic classifications with slugs), related (the five episodes from other shows most related to this one — slug, show, score, band). Request only what you need. Unknown values are rejected with invalid_parameter.
transcript_format	string	no	text	Transcript format for include: ["transcript"]. One of dialogue, text, srt.
transcript_speaker	string	no	—	Filter the transcript to one speaker (name or entity slug).
transcript_start	number	no	—	Transcript start clip in seconds.
transcript_end	number	no	—	Transcript end clip in seconds.

The old boolean flags include_transcript, include_full_segments, and include_full_entities are gone — pass the matching value in the include array instead (["transcript"], ["segments"], ["entities"]). The transcript_* sub-parameters are unchanged.

Output

A markdown document with the episode title as an H2, then KV lines (**Podcast:** as Title (slug), **Published:**, **Episode slug:**, **Duration:**, **Counts:** summarising segments/clips/entities), then a paragraph description. The following H3 sections appear when the underlying data is present:

• ### Speakers — bullets of Name (role) (slug appended as — slug: <entity_slug> when linked).
• ### Top entities — bullets of Name (slug) — salience X.XXX, N occurrences. Capped at 20 unless include contains entities.
• ### Topics — bullets of Name — slug: <topic_slug>. Only when include contains topics. The slug round-trips into particle_topic_browse and topic_slug filters.
• ### Segments — only when include contains segments. Bullets formatted #N TYPE — Title @ Xs–Ys (transcript lines A–B). Segments are contiguous in those line numbers, not in seconds: where the suffix is present, the gap between one segment’s end and the next’s start is silence with no transcribed speech. The suffix is omitted on episodes segmented before line ranges were recorded, and a small share of those do leave lines uncovered, so the silence guarantee does not extend to them.
• ### Clips — only when include contains clips. Bullets formatted Title (clip-id) — TYPE, engagement N @ Xs–Ys.
• ## Related episodes for <episode-slug> (N) — only when include contains related. Numbered lines formatted **Episode title** (episode-slug) — Podcast title (podcast-slug) · YYYY-MM-DD — score 0.84, strong, best first, from other shows only; a single sentence instead when no related episodes have been computed yet. The JSON form carries the section as related (episodes plus a note when empty), so an honoured-but-empty include is distinguishable from one not requested.
• ### Transcript (<format>) — only when include contains transcript. Body is the raw transcript text/SRT, or for dialogue format a series of **Speaker:** text lines.

Sample (episode_slug="why-are-we-all-so-stressed", include=["segments","topics"], truncated):

## Why are we all so stressed?

**Podcast:** 6 Minute English (6-minute-english)
**Published:** 2026-04-23
**Episode slug:** why-are-we-all-so-stressed
**Duration:** 402s
**Counts:** 9 segments, 0 clips, 4 entities
Worrying about exams, work or climate change in the news. …

### Speakers

- Neil (HOST)
- Becca (HOST)

### Top entities

- Claudia Hammond (claudia-hammond) — salience 0.875, 15 occurrences
- English (english) — salience 0.375, 11 occurrences

### Topics

- Mental Health — slug: society/mental-health
- Stress Management — slug: society/mental-health/stress-management

### Segments

- #1 INTRO — Show and series introduction @ 0s–41s
- #2 PERSONAL_BANTER — Hosts share what makes them stressed @ 42s–94s
- #3 TOPIC_DISCUSSION — Defining overwhelm and burnout @ 95s–150s

Example

Agent calls: particle_podcast_get_episode {
  "episode_slug": "whcd-shooting-aftermath-musk-and-altman-face-off",
  "include": ["segments", "transcript"],
  "transcript_format": "dialogue",
  "transcript_speaker": "kara-swisher"
}

Pulling a known episode’s full highlight-clip list:

Agent calls: particle_podcast_get_episode {
  "episode_slug": "all-in-200",
  "include": ["clips"]
}

Related

• REST equivalents: GET /v1/podcasts/episodes/{id}, /transcript, /segments, /entities.
• For dialogue around a specific entity in this episode, use particle_podcast_find_mentions with episode_slug.
• For ranked dialogue across the catalog, use particle_podcast_search_transcripts.
• For the full ranked list of related episodes with the reasons behind each match, a recency window, or paging, use particle_podcast_list_related_episodes; include: ["related"] here is the five-row inline form.