Back to search
35
MCP MCP 2025-11-25 http

BibiGPT AI video & audio summarizer

mcp.so:jimmylv/bibigpt-skill
Watch this dependency Share evidence report
Uptime
—
no probes in 30d
Response
722ms
earlier probe
Tools
30
callable
Resources
0
readable
Prompts
0
available
Primary URL
https://bibigpt.co/api/mcp
Score: 35/100
The server exists but requires sign-in before we can list its tools.
partial Answers the MCP handshake 35/50
The endpoint answered HTTP 401 to an anonymous initialize, so it exists, but we could not list its tools without credentials. It does advertise OAuth protected-resource metadata.
How to earn +15 points
Let anonymous clients discover what the server offers
Answer initialize and tools/list without credentials and require auth only on tools/call. If the whole server must stay private, return a WWW-Authenticate header with resource_metadata pointing at /.well-known/oauth-protected-resource (RFC 9728) so MCP clients can sign in on their own; 2026-07-28 clients identify themselves with a Client ID Metadata Document rather than dynamic registration.
Docs →
n/a Lists tools, resources or prompts 0/30
Not scored: a catalogue only counts once the handshake succeeds.
n/a 3 or more catalogue items 0/10
Not scored: a catalogue only counts once the handshake succeeds.
fail Describes itself 0/10
0 characters; more than 30 earns these points.
How to earn +10 points
Describe the server in one or two sentences
Set description in your server.json (or the registry listing) to say what the server does and for whom.

Tools · 30

summarize_video

Summarize an online video or podcast. Supports YouTube, Bilibili, TikTok, Twitter/X, Xiaohongshu, podcasts, and more. Returns AI-generated summary text.

summarize_video_with_config

Summarize a video with custom prompt, AI model, and output language. Use this when the user wants a specific summary style or language.

summarize_by_chapter

Summarize a video chapter by chapter. Returns structured summaries for each section/chapter of the video.

create_summary_task

Create an async summary task for a video. Returns a taskId to poll with get_task_status. Use for long videos that take time to process.

get_task_status

Check the status of an async summary task. Returns the summary when completed, or current status if still processing.

get_subtitle

Get the full transcript/subtitles of a video or podcast. Returns subtitle array with timestamps in the detail field.

get_account_info

Return the current authenticated user's account info: userId, email, plan tier, member expiration, and remaining summarization minutes.

get_latest_feed

Pull the user's personal feed from the persisted channel_videos catalog (keyset pagination). Items the user has already summarized include a non-null contentId.

get_channel_health

Check RSS reachability for every subscribed channel. Returns per-channel ok/error and the latest published_at when available. Reuses an in-process cache; calling shortly after get_latest_feed is essen…

mark_feed_seen

Mark the feed as "read up to here" by bumping the per-(user, channel) last_seen_at. Without channelUrl, bumps every subscription. Use after presenting feed items so the next get_latest_feed call retur…

list_saved_videos

List the user's saved/summarized videos with pagination. Optional filters: channelId, tag, smartFolder, highlighted only, sort order.

get_saved_video

Get a single saved video with everything attached: title, summary, user note, chapter summaries, subtitle array, source URL.

search_saved_videos

Semantic + keyword search across the user's saved video library. Returns matched snippets with timestamps so the agent can deep-link into specific moments.

list_channels

List the channels the user subscribes to, with last-summarized-at timestamps. Useful for the agent to surface recent activity.

subscribe_channel

Subscribe to a YouTube/Bilibili/podcast channel by URL. Subject to subscription quota (Pro/Plus higher limits).

unsubscribe_channel

Unsubscribe from a previously subscribed channel by its URL.

get_channel_videos

Get the latest videos for a channel by URL (uses RSS / AIRSS). Use before subscribing to preview, or to refresh a channel's recent uploads.

list_collections

List the user's collections. `scope` controls whether to return owned, purchased, or both.

get_collection

Get a collection's details including its items (saved videos), aggregated summary, and citations if present.

create_collection

Create a new collection. Pass `isPublic: true` to make it shareable.

get_collection_chat_history

Read prior chat messages and AI-suggested questions for a collection (cross-video RAG conversation). Returns empty arrays if the user has not chatted yet.

add_to_collection

Add an item (saved video) to a collection by collectionId + sourceUrl or contentId.

list_notes

List all the user's saved notes (their personalized summaries) across content. Cursor-paginated by updated_at desc.

get_note

Get the user's note (personalized summary) for a specific saved video by contentId.

update_note

Save or update the user's note (personalized summary) for a saved video. Overwrites any prior note for the same contentId.

generate_video_mindmap

Generate an XMind mindmap file (.xmind) from a saved video summary. Returns a downloadable file URL. Use after the video is summarized.

extract_video_visuals

Create an async visual-analysis task for a video URL: PPT slide extraction + subtitle OCR + scene markdown. Returns a taskId; poll vision.getVideoProcessingTask for status & artifacts (slides PDF, mar…

generate_summary_by_prompt

Generate a custom summary for a saved video using a user-provided prompt. Useful for re-angling existing content (e.g., "extract top 3 actionable insights"). Result is cached per (contentId, prompt) s…

get_notion_status

Check whether the user has connected their Notion workspace to BibiGPT. Returns connection state and workspace metadata; use before calling export_to_notion to give a clear error if not connected.

export_to_notion

Export a saved video summary (note + video metadata) to the user's connected Notion workspace as a new page. Requires a prior Notion OAuth connection (see get_notion_status).

How to use

Claude Desktop or claude.ai: add a custom connector (Customize → Connectors; on Team and Enterprise plans an owner adds it under Organization settings → Connectors) with this server URL: https://bibigpt.co/api/mcp

Cursor or Cline: add to your MCP config:

{
  "mcpServers": {
    "bibigpt_ai_video_\u0026_audio_summarizer": {
      "url": "https://bibigpt.co/api/mcp",
      "transport": "http"
    }
  }
}