CleoLM
Server Details
Search and cite timestamped YouTube and podcast transcripts; transcribe any video or episode.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 10 tools
Most tools have distinct roles: discovery, ingestion, job status, retrieval, and presentation are separated cleanly. A few boundaries are soft—discover vs list_episodes for finding episodes, and search vs get_passage/get_transcript for reading transcripts—but the descriptions provide guidance.
All names use snake_case, but the conventions are mixed: some are bare verbs or nouns (account, discover, ingest, search, show_player) while others use get_/list_ prefixes. The set is readable and predictable in practice, but it does not follow a single verb_noun pattern.
Ten tools is well-scoped for a transcription/search service: account, discovery, ingestion, job polling, retrieval, and display each earn a place. There is no obvious tool bloat or critical missing area that would require more tools.
The core lifecycle is covered: find media, ingest it, poll jobs, search, read passages/transcripts, and present video answers. Minor gaps remain around managing the user's library—such as listing or deleting ingested items—and playback is YouTube-only, but core workflows are supported.
Available Tools
10 toolsaccountPlan and creditsARead-onlyIdempotentInspect
The connected CleoLM account: plan, credits remaining this period, reset date, and limits. Free. Call this when a tool reports credits_exhausted or upgrade_required.
Inputs: no named parameters. Returns: plan, limits, credits, free_tier, +4 more
discoverFind videos and podcastsARead-onlyIdempotentInspect
Find YouTube videos or podcasts on the open web (not limited to the indexed channels). type "youtube" returns videos; "podcast" returns shows with their RSS feed; "podcast_episode" returns individual episodes. Spotify-listed shows are found through their public RSS feed (Spotify-exclusive shows are not available). Next: ingest a result to transcribe it, or list_episodes on a show/channel.
Inputs: query (required), type, limit. Returns: type, results
get_jobCheck an ingest jobARead-onlyIdempotentInspect
Status of an ingest job (queued, running, completed, failed). Poll every 15-30 seconds until completed. Free.
Inputs: job_id (required). Returns: url, next, error, job_id, +4 more
get_passageGet transcript around a timestampARead-onlyIdempotentInspect
The transcript window around a moment in an item — use after search to read the full context of a hit (e.g. 90 seconds before and after). Free. For a summary or key points of one specific YouTube video, call show_player first with a short summary and the key points as moments, then write your answer citing timestamps.
Inputs: at_seconds (required), id, url, after_seconds, +1 more. Returns: item, segments
get_transcriptGet a transcriptARead-onlyIdempotentInspect
Timestamped transcript of an indexed or ingested item, paginated (~30k characters per page). If it returns not_ingested, call ingest first. Prefer search/get_passage when you only need part of it. Costs 0.2 credits per page. For a summary or key points of one specific YouTube video, call show_player first with a short summary and the key points as moments, then write your answer citing timestamps.
Inputs: id, url, cursor, format, +1 more. Returns: item, page, text, quality, +1 more
ingestTranscribe and index mediaAIdempotentInspect
Transcribe a YouTube video, podcast episode (Apple Podcasts, Spotify-with-RSS, RSS, Buzzsprout, Libsyn, etc.) and add it to the user's searchable library. Already-transcribed media returns instantly at no cost. Otherwise it runs as a job: waits up to wait_seconds, then returns status "queued"/"running" — poll get_job. quality "premium" re-transcribes with high-accuracy ASR when the account's plan includes it (best for multi-speaker interviews). Charged by route: captions 0.5 credits, ASR 20/hr, AI video 36/hr, podcasts 10/hr; worst case is reserved up front and the difference refunded.
Inputs: url (required), quality, wait_seconds. Returns: url, next, error, job_id, +4 more
list_channelsList indexed channelsARead-onlyIdempotentInspect
Channels in the continuously-updated public index that search covers. Optional query filters by name. Free.
Inputs: query. Returns: total, channels
list_episodesList channel or show episodesARead-onlyIdempotentInspect
Latest uploads of a YouTube channel (handle like "@lexfridman", channel URL, or UC… id) or episodes of a podcast (RSS feed URL, Apple Podcasts show URL, or Spotify show URL). Newest first. Pass an episode's url to ingest.
Inputs: source (required), limit. Returns: source, episodes
searchSearch transcriptsARead-onlyIdempotentInspect
Semantic + keyword search across transcripts of 158 continuously-updated YouTube/podcast channels (interviews, tech, business, science, health, politics) and the user's own ingested library. Returns the most relevant passages with timestamps and deep links. Use this first for "what has X said about Y" questions; then call get_passage for more context around a hit, or get_transcript for the full text. Call list_channels to see which channels are indexed. For a summary or key points of one specific YouTube video, call show_player first with a short summary and the key points as moments, then write your answer citing timestamps.
Inputs: query (required), limit, scope, channels, +3 more. Returns: query, results
show_playerShow video playerARead-onlyIdempotentInspect
Show your answer about one YouTube video as an interactive card: the video player, a one- or two-sentence summary, and your answer organized as sections (a short heading each, with points that each carry the seconds they come from; clicking a point plays that moment). Use it for any answer about one specific video (a summary, brief, key points, quotes, or a question about what it says). Call it BEFORE writing your reply: the card is the organized answer, so your reply should then be only a few sentences answering the question directly, without repeating the points. Check each point's seconds against the transcript first and call it once per answer: each call shows a separate card in the chat. It returns CleoLM links for each point (watch_url) for any timestamp you mention. moments (a flat list) is the older form. Free. YouTube only; not for podcasts.
Inputs: id, url, title, channel, +3 more. Returns: url, title, channel, item_id, +5 more
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Search and query the transcripts of YouTube expert channels you follow, with cited answers.
Search 4M+ podcasts & YouTube, transcribe any episode, search transcripts, generate AI lessons.
Transcribe audio, video and YouTube links, then search, summarize and export your transcripts.
Transcripts, translations, summaries, subtitles, and search for public YouTube videos.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables AI agents to search a private, timestamped YouTube transcript library and retrieve cited passages with video links and exact timestamps. Supports hybrid vector and full-text search across entire channels so agents can answer from trusted sources rather than rewatching videos.1MIT
- AlicenseAqualityCmaintenanceEnables AI assistants to fetch YouTube transcripts with timestamps, search within a video, and keep a local searchable memory of every transcript fetched.4MIT
- AlicenseNot gradedqualityBmaintenanceEnables search across videos you've watched via transcripts, on-screen text, and frames, citing exact timestamps. Point it at videos, channels, or playlists; it indexes everything locally and answers queries with deep links to the exact second.3MIT
- AlicenseAqualityBmaintenanceAllows AI agents to fetch, search, and summarize YouTube video transcripts without an API key, including chapter summaries.4MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.