An MCP server that lets any AI client read public Bilibili data: video info, subtitles, danmaku, danmaku hotwords, comments, and search.
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"bilibili-data-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# bilibili-mcp An MCP server that lets any AI client (Claude, Cursor, ...) read **public** Bilibili data: video info, subtitles, danmaku, and search. It's a thin wrapper over the mature [`bilibili-api`](https://github.com/Nemo2011/bilibili-api) library — all WBI signing, cookies, and anti-abuse handling are delegated to it. > Reads public data only. Video info and search work as a guest (no login). Subtitles are gated behind login by Bilibili itself, so the subtitle tool needs your own SESSDATA cookie (see below). For personal/research use at a reasonable request rate. ## Tools | Tool | Description | | --- | --- | | `get_video_info(bvid)` | Title, uploader, play/like/coin counts, duration, publish date, description. Accepts a BV id or a full video URL. **Guest, no login.** | | `get_video_subtitle(bvid, lang="zh-CN", with_timestamp=False)` | Full subtitle/transcript text — great for "summarize this video". **Requires login** (`BILI_SESSDATA`, see below); returns clear guidance if no cookie is set, and a clear message when a video simply has no subtitles. | | `get_video_danmaku(bvid, limit=200, with_timestamp=False)` | All danmaku (弹幕, on-screen bullet comments), ordered by appearance time — great for asking the AI what memes viewers spam, the overall mood, or which moments are "高能". **Guest, no login.** | | `get_danmaku_hotwords(bvid, top=20, segments=5)` | Aggregates danmaku into a top repeated-comment ranking plus the highest-density "高能" time segments — a quick read on what viewers spam and which moments spike. **Guest, no login.** | | `get_video_comments(bvid, sort="hot", limit=20)` | Top-level video comments (username, likes, content), sorted by likes (`hot`) or newest (`time`) — great for summarizing reception or finding debate. **Guest, no login.** | | `search_videos(keyword, page=1, limit=10)` | Search videos by keyword. **Guest, no login.** | ## Install The PyPI package is **`bilibili-data-mcp`** (the import module is `bilibili_mcp`). ### Option A — …
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.