AI AGENT SKILLS

Bilibili All In One

一个面向 Design 场景的 Agent 技能。原始说明：A comprehensive Bilibili toolkit that integrates hot trending monitoring, video downloading, video watching/playback, subtitle downloading, and video publish...

下载技能包打开来源页 Design

SKILL.md

skill.md

name: bilibili-all-in-one
description: >
A comprehensive Bilibili toolkit that integrates hot trending monitoring,
video downloading, video watching/playback, subtitle downloading,
and video publishing capabilities into a single unified skill.
Supports Bilibili session cookie authentication for publishing and
high-quality downloads. Requests go to official Bilibili API endpoints
over HTTPS.
version: 1.0.18
type: code
implementation: python
interface: cli-and-api
runtime: python>=3.8
languages:

zh-CN
en

tags:

bilibili
video-download
hot-trending
subtitle
danmaku
video-publish
video-player

batch-download
multi-format

author: wscats
license: MIT
homepage: https://github.com/wscats/bilibili-all-in-one
repository: https://github.com/wscats/bilibili-all-in-one
entry_point: main.py
requiredenvvars: []
optionalenvvars:

BILIBILI_SESSDATA
BILIBILIBILIJCT
BILIBILI_BUVID3
BILIBILI_PERSIST

install: pip install -r requirements.txt

Bilibili All-in-One Skill

A comprehensive Bilibili toolkit that integrates hot trending monitoring, video downloading, video watching/playback, subtitle downloading, and video publishing capabilities into a single unified skill.

⚠️ Optional Environment Variables: BILIBILI_SESSDATA, BILIBILI_BILI_JCT (optional), BILIBILI_BUVID3 (optional), BILIBILI_PERSIST (optional)

These are sensitive Bilibili session cookies needed only for publishing and high-quality (1080p+/4K) downloads.

Most features work WITHOUT any credentials: hot monitoring, standard-quality downloads, subtitle listing, danmaku, stats viewing.

📦 Install: pip install -r requirements.txt (all standard PyPI packages: httpx, aiohttp, beautifulsoup4, lxml, requests)

🔗 Source: github.com/wscats/bilibili-all-in-one

何时激活

当用户明确请求以下 Bilibili 相关操作时，本 Skill 可被激活：

| 触发场景 | 匹配的模块 | 典型触发词 |
|---|---|---|
| 查看B站热门、热搜、排行榜、必看榜 | 🔥 Hot Monitor | "热门"、"热搜"、"排行"、"趋势"、"必看"、"流行"、"榜单" |
| 下载B站视频、提取音频、批量下载 | ⬇️ Downloader | "下载"、"保存视频"、"提取音频"、"导出MP4"、"批量下载" |
| 查看视频播放量、点赞数、数据追踪、对比 | 👀 Watcher | "播放量"、"点赞"、"数据"、"统计"、"对比"、"监控"、"追踪"、"观看量" |
| 下载字幕、转换字幕格式、合并字幕 | 📝 Subtitle | "字幕"、"CC"、"SRT"、"ASS"、"字幕下载"、"字幕转换"、"翻译" |
| 播放视频、获取弹幕、播放列表 | ▶️ Player | "播放"、"弹幕"、"播放地址"、"分P"、"播放列表"、"danmaku" |
| 上传视频、发布、定时发布、草稿、编辑 | 📤 Publisher | "上传"、"发布"、"投稿"、"定时发布"、"草稿"、"编辑视频" |

⚠️ 注意：本 Skill 不会仅因消息中出现 Bilibili 链接或 BV 号就自动激活。只有当用户明确表达了操作意图（如"下载这个视频"、"查看热门"等）时才会被调用。涉及写操作（发布/编辑）时，需要用户显式提供凭证。

Features

| Module | Description |
|---|---|
| 🔥 Hot Monitor | Monitor Bilibili hot/trending videos and topics in real-time |
| ⬇️ Downloader | Download Bilibili videos with multiple quality and format options |
| 👀 Watcher | Watch and track video engagement metrics (supports Bilibili) |
| 📝 Subtitle | Download and process subtitles in multiple formats and languages |
| ▶️ Player | Get playback URLs, danmaku (bullet comments), and playlist info |
| 📤 Publisher | Upload, schedule, edit, and manage videos on Bilibili |

Installation

Requirements

Python >= 3.8
ffmpeg (optional, for merging video/audio streams)

Install Dependencies

pip install -r requirements.txt

Dependencies

httpx >= 0.24.0
aiohttp >= 3.8.0
beautifulsoup4 >= 4.12.0
lxml >= 4.9.0
requests >= 2.31.0
faster-whisper >= 1.0.0 (optional, for speech recognition subtitle fallback)

Configuration

Some features (downloading high-quality videos, publishing, etc.) require Bilibili authentication. You can provide credentials in three ways:

1. Environment Variables

export BILIBILI_SESSDATA="your_sessdata"
export BILIBILI_BILI_JCT="your_bili_jct"
export BILIBILI_BUVID3="your_buvid3"

2. Credential File

Create a JSON file (e.g., credentials.json):

{
  "sessdata": "your_sessdata",
  "bili_jct": "your_bili_jct",
  "buvid3": "your_buvid3"
}

3. Direct Parameters

Pass credentials directly when initializing:

from main import BilibiliAllInOne

app = BilibiliAllInOne(
    sessdata="your_sessdata",
    bili_jct="your_bili_jct",
    buvid3="your_buvid3",
)

4. Persistent Storage (Optional)

By default, credentials are kept in-memory only and are not saved to disk. To enable automatic persistence across sessions:

# Via environment variable
export BILIBILI_PERSIST=1

# Or via code
app = BilibiliAllInOne(persist=True)

When persistence is enabled:

Credentials are auto-saved to .credentials.json (with 0600 permissions) after initialization
On next startup, credentials are auto-loaded from this file
You can toggle persistence at runtime: app.auth.persist = True / app.auth.persist = False
To delete the persisted file: app.auth.clear_persisted()

How to get cookies: Log in to bilibili.com, open browser DevTools (F12) → Application → Cookies, and copy the values of SESSDATA, bili_jct, and buvid3.

⚠️ Security & Privacy

Credential Handling

This skill handles sensitive Bilibili session cookies. Please read the following carefully:

| Concern | Detail |
|---|---|
| What credentials are needed? | SESSDATA, bili_jct, buvid3 — Bilibili full browser session cookies (not limited API keys). Providing them grants broad access to your Bilibili account. |
| Which features require authentication? | Publishing (upload/edit/schedule/draft), downloading 1080p+/4K quality videos |
| Which features work WITHOUT credentials? | Hot monitoring, standard-quality downloads, subtitle listing, danmaku fetching, stats viewing |
| Where are credentials sent? | To official Bilibili API endpoints (api.bilibili.com, member.bilibili.com) over HTTPS only |
| Are credentials persisted to disk? | NO by default — credentials stay in memory. Set persist=True or BILIBILI_PERSIST=1 to opt-in to automatic persistence (saved to .credentials.json with 0600 permissions). You can also manually call auth.save_to_file() |
| File permissions for saved credentials | 0600 (owner read/write only) — restrictive by default |

Best Practices

🧪 Use a test account — Do NOT provide your primary Bilibili account cookies for evaluation/testing purposes. These are full session cookies that grant broad account access (not limited API keys).
🔒 Prefer in-memory credentials — Pass credentials via environment variables or direct parameters rather than saving to a file. Only enable persist=True if you need credentials to survive across sessions.
📁 If you enable persistence — Credentials are saved with 0600 permissions. Use auth.clear_persisted() or auth.persist = False to remove the file when no longer needed.
🐳 Run in isolation — When possible, run this skill in an isolated container/environment and inspect network traffic.
🌐 Verify network traffic — All HTTP requests go to Bilibili's official domains only. You can verify by monitoring outbound connections.
❌ No exfiltration — This skill does NOT send credentials to any third-party service, analytics endpoint, or telemetry server.
🔑 Credential scope — SESSDATA and bili_jct are full session cookies. They are NOT scoped/limited API keys. Treat them with the same care as your account password.

Network Endpoints Used

| Domain | Purpose |
|---|---|
| api.bilibili.com | Video info, stats, hot lists, subtitles, danmaku, playback URLs |
| member.bilibili.com | Video publishing (upload, edit) |
| upos-sz-upcdnbda2.bilivideo.com | Video file upload CDN |
| www.bilibili.com | Web page scraping fallback |

Credential Requirement by Module

| Module | Auth Required? | Notes |
|---|---|---|
| 🔥 Hot Monitor | ❌ No | All public APIs |
| ⬇️ Downloader | ⚠️ Optional | Required only for 1080p+ / 4K quality |
| 👀 Watcher | ❌ No | Public stats APIs |
| 📝 Subtitle | ❌ No | Public subtitle APIs |
| ▶️ Player | ⚠️ Optional | Required for high-quality playback URLs |
| 📤 Publisher | ✅ Required | All operations need SESSDATA + bili_jct |

Usage

CLI

python main.py <skill_name> <action> [params_json]

Python API

import asyncio
from main import BilibiliAllInOne

app = BilibiliAllInOne()

async def demo():
    result = await app.execute("hot_monitor", "get_hot", limit=5)
    print(result)

asyncio.run(demo())

Skills Reference

1. 🔥 Hot Monitor (`bilibili_hot_monitor`)

Monitor Bilibili hot/trending videos and topics in real-time. Supports filtering by category, tracking rank changes.

Actions

| Action | Description | Parameters |
|---|---|---|
| get_hot | Get popular/hot videos | page, page_size |
| get_trending | Get trending series/topics | limit |
| get_weekly | Get weekly must-watch list | number (week number, optional) |
| get_rank | Get category ranking videos | category, limit |

Supported Categories

all, anime, music, dance, game, tech, life, food, car, fashion, entertainment, movie, tv

Examples

# Get top 10 hot videos
python main.py hot_monitor get_hot '{"page_size": 10}'

# Get trending topics
python main.py hot_monitor get_trending '{"limit": 5}'

# Get this week's must-watch
python main.py hot_monitor get_weekly

# Get game category rankings
python main.py hot_monitor get_rank '{"category": "game", "limit": 10}'

# Python API
result = await app.execute("hot_monitor", "get_hot", page_size=10)
result = await app.execute("hot_monitor", "get_rank", category="game", limit=10)

2. ⬇️ Downloader (`bilibili_downloader`)

Download Bilibili videos with support for multiple quality options, batch downloading, and format selection.

Actions

| Action | Description | Parameters |
|---|---|---|
| get_info | Get video information | url |
| get_formats | List available qualities/formats | url |
| download | Download a single video | url, quality, output_dir, format, page |
| batch_download | Download multiple videos | urls, quality, output_dir, format |

Quality Options

360p, 480p, 720p, 1080p (default), 1080p+, 4k

Format Options

mp4 (default), flv, mp3 (audio only)

Examples

# Get video info
python main.py downloader get_info '{"url": "BV1xx411c7mD"}'

# List available formats
python main.py downloader get_formats '{"url": "BV1xx411c7mD"}'

# Download in 1080p MP4
python main.py downloader download '{"url": "BV1xx411c7mD", "quality": "1080p", "format": "mp4"}'

# Extract audio only
python main.py downloader download '{"url": "BV1xx411c7mD", "format": "mp3"}'

# Batch download
python main.py downloader batch_download '{"urls": ["BV1xx411c7mD", "BV1yy411c8nE"], "quality": "720p"}'

# Python API
info = await app.execute("downloader", "get_info", url="BV1xx411c7mD")
result = await app.execute("downloader", "download", url="BV1xx411c7mD", quality="1080p")

3. 👀 Watcher (`bilibili_watcher`)

Watch and monitor Bilibili videos. Track view counts, comments, likes, and other engagement metrics over time.

Actions

| Action | Description | Parameters |
|---|---|---|
| watch | Get detailed video information | url |
| get_stats | Get current engagement statistics | url |
| track | Track metrics over time | url, interval (minutes), duration (hours) |
| compare | Compare multiple videos | urls |

Supported Platforms

Bilibili: https://www.bilibili.com/video/BVxxxxxx or BVxxxxxx

Examples

# Get video details
python main.py watcher watch '{"url": "BV1xx411c7mD"}'

# Get current stats
python main.py watcher get_stats '{"url": "BV1xx411c7mD"}'

# Track views every 30 minutes for 12 hours
python main.py watcher track '{"url": "BV1xx411c7mD", "interval": 30, "duration": 12}'

# Compare multiple videos
python main.py watcher compare '{"urls": ["BV1xx411c7mD", "BV1yy411c8nE"]}'

# Python API

comparison = await app.execute("watcher", "compare", urls=["BV1xx411c7mD", "BV1yy411c8nE"])

4. 📝 Subtitle (`bilibili_subtitle`)

Download and process subtitles/CC from Bilibili videos. Supports multiple subtitle formats and languages.

When no CC subtitles are available, the module automatically falls back to:

Speech Recognition — Downloads the video's audio and transcribes it using faster-whisper (requires pip install faster-whisper)
Danmaku Extraction — Fetches bullet comments from the video as a text reference

Both fallback results are returned together when triggered.

Actions

| Action | Description | Parameters |
|---|---|---|
| list | List available subtitles | url |
| download | Download subtitles (with auto-fallback) | url, language, format, output_dir |
| convert | Convert subtitle format | input_path, output_format, output_dir |
| merge | Merge multiple subtitle files | input_paths, output_path, output_format |

Supported Formats

srt (default), ass, vtt, txt, json

Supported Languages

zh-CN (default), en, ja, and other language codes available on the video.

Fallback Strategy

When download is called and no CC subtitles exist:

CC Subtitle Available? ──Yes──▶ Download CC subtitle
        │
       No
        │
        ▼
┌─────────────────────────────────────┐
│  Fallback 1: Speech Recognition     │
│  Download audio → faster-whisper    │
│  Output: {title}_transcribed.srt    │
├─────────────────────────────────────┤
│  Fallback 2: Danmaku Extraction     │
│  Fetch bullet comments → SRT        │
│  Output: {title}_danmaku.srt        │
└─────────────────────────────────────┘

Examples

# List available subtitles
python main.py subtitle list '{"url": "BV1xx411c7mD"}'

# Download Chinese subtitles in SRT format (auto-fallback if no CC)
python main.py subtitle download '{"url": "BV1xx411c7mD", "language": "zh-CN", "format": "srt"}'

# Download English subtitles in ASS format
python main.py subtitle download '{"url": "BV1xx411c7mD", "language": "en", "format": "ass"}'

# Convert SRT to VTT
python main.py subtitle convert '{"input_path": "./subtitles/video.srt", "output_format": "vtt"}'

# Merge subtitle files
python main.py subtitle merge '{"input_paths": ["part1.srt", "part2.srt"], "output_path": "merged.srt"}'

# Python API
subs = await app.execute("subtitle", "list", url="BV1xx411c7mD")
result = await app.execute("subtitle", "download", url="BV1xx411c7mD", language="zh-CN", format="srt")

# When no CC subtitles, result includes both fallback outputs:
# result["transcription"]["filepath"]  → speech recognition SRT
# result["danmaku"]["filepath"]        → danmaku SRT

5. ▶️ Player (`bilibili_player`)

Play Bilibili videos with support for playback control, playlist management, and danmaku (bullet comments) display.

Actions

| Action | Description | Parameters |
|---|---|---|
| play | Get complete playback info | url, quality, page |
| get_playurl | Get direct play URLs | url, quality, page |
| get_danmaku | Get danmaku/bullet comments | url, page, segment |
| get_playlist | Get playlist/multi-part info | url |

Danmaku Modes

| Mode | Description |
|---|---|
| 1 | Scroll (right to left) |
| 4 | Bottom fixed |
| 5 | Top fixed |

Examples

# Get playback info
python main.py player play '{"url": "BV1xx411c7mD", "quality": "1080p"}'

# Get direct play URLs
python main.py player get_playurl '{"url": "BV1xx411c7mD", "quality": "720p"}'

# Get danmaku
python main.py player get_danmaku '{"url": "BV1xx411c7mD"}'

# Get playlist for multi-part video
python main.py player get_playlist '{"url": "BV1xx411c7mD"}'

# Get page 3 of a multi-part video
python main.py player play '{"url": "BV1xx411c7mD", "quality": "1080p", "page": 3}'

# Python API
play_info = await app.execute("player", "play", url="BV1xx411c7mD", quality="1080p")
danmaku = await app.execute("player", "get_danmaku", url="BV1xx411c7mD")
playlist = await app.execute("player", "get_playlist", url="BV1xx411c7mD")

6. 📤 Publisher (`bilibili_publisher`)

Publish videos to Bilibili. Supports uploading videos, setting metadata, scheduling publications, and managing drafts.

⚠️ Authentication Required: All publisher actions require valid Bilibili credentials.

Actions

| Action | Description | Parameters |
|---|---|---|
| upload | Upload and publish a video | file_path, title, description, tags, category, cover_path, dynamic, no_reprint, open_elec |
| draft | Save as draft | file_path, title, description, tags, category, cover_path |
| schedule | Schedule future publication | file_path, title, schedule_time, description, tags, category, cover_path |
| edit | Edit existing video metadata | bvid, file_path, title, description, tags, cover_path |

Upload Parameters

| Parameter | Type | Default | Description |
|---|---|---|---|
| file_path | string | required | Path to the video file |
| title | string | required | Video title (max 80 chars) |
| description | string | "" | Video description (max 2000 chars) |
| tags | string[] | ["bilibili"] | Tags (max 12, each max 20 chars) |
| category | string | "171" | Category TID |
| cover_path | string | null | Path to cover image (JPG/PNG) |
| no_reprint | int | 1 | 1 = original content, 0 = repost |
| open_elec | int | 0 | 1 = enable charging, 0 = disable |

Examples

# Upload and publish
python main.py publisher upload '{"file_path": "./video.mp4", "title": "My Video", "description": "Hello World", "tags": ["test", "demo"], "category": "171"}'

# Save as draft
python main.py publisher draft '{"file_path": "./video.mp4", "title": "Draft Video"}'

# Schedule publication
python main.py publisher schedule '{"file_path": "./video.mp4", "title": "Scheduled Video", "schedule_time": "2025-12-31T20:00:00+08:00"}'

# Edit video metadata (requires re-uploading the video file)
python main.py publisher edit '{"bvid": "BV1xx411c7mD", "file_path": "./video.mp4", "title": "New Title", "tags": ["updated"]}'

# Python API (authentication required)
app = BilibiliAllInOne(sessdata="xxx", bili_jct="xxx", buvid3="xxx")

result = await app.execute("publisher", "upload",
    file_path="./video.mp4",
    title="My Video",
    description="Published via bilibili-all-in-one",
    tags=["python", "bilibili"],
)

# Edit video (requires file_path for re-upload)
result = await app.execute("publisher", "edit",
    bvid="BV1xx411c7mD",
    file_path="./video.mp4",
    title="New Title",
    tags=["updated"],
)

Project Structure

bilibili-all-in-one/
├── skill.json              # Skill configuration & parameter schema
├── skill.md                # This documentation file
├── README.md               # Project README (Chinese)
├── LICENSE                  # MIT License
├── requirements.txt        # Python dependencies
├── .gitignore              # Git ignore rules
├── main.py                 # Entry point & unified BilibiliAllInOne class
└── src/
    ├── __init__.py         # Package exports
    ├── auth.py             # Authentication & credential management
    ├── utils.py            # Shared utilities, API constants, helpers
    ├── hot_monitor.py      # Hot/trending video monitoring
    ├── downloader.py       # Video downloading
    ├── watcher.py          # Video watching & stats tracking
    ├── subtitle.py         # Subtitle downloading & processing
    ├── player.py           # Video playback & danmaku
    └── publisher.py        # Video uploading & publishing

Response Format

All skill actions return a JSON object with a unified structure:

{
  "success": true,
  "...": "action-specific fields"
}

On error:

{
  "success": false,
  "message": "Error description"
}

License

MIT

Design

Humanizer

一个面向 Design 场景的 Agent 技能。原始说明：Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflat...

Design 低风险

skill.md

Bilibili All-in-One Skill

何时激活

Features

Installation

Requirements

Install Dependencies

Dependencies

Configuration

1. Environment Variables

2. Credential File

3. Direct Parameters

4. Persistent Storage (Optional)

⚠️ Security & Privacy

Credential Handling

Best Practices

Network Endpoints Used

Credential Requirement by Module

Usage

CLI

Python API

Skills Reference

1. 🔥 Hot Monitor (bilibili_hot_monitor)

Actions

Supported Categories

Examples

2. ⬇️ Downloader (bilibili_downloader)

Actions

Quality Options

Format Options

Examples

3. 👀 Watcher (bilibili_watcher)

Actions

Supported Platforms

Examples

4. 📝 Subtitle (bilibili_subtitle)

Actions

Supported Formats

Supported Languages

Fallback Strategy

Examples

5. ▶️ Player (bilibili_player)

Actions

Danmaku Modes

Examples

6. 📤 Publisher (bilibili_publisher)

Actions

Upload Parameters

Examples

Project Structure

Response Format

License

相关技能

Humanizer

Nano Banana Pro

Free Ride - Unlimited free AI

Caldav Calendar

SuperDesign

Intelligent Stocks Screener

1. 🔥 Hot Monitor (`bilibili_hot_monitor`)

2. ⬇️ Downloader (`bilibili_downloader`)

3. 👀 Watcher (`bilibili_watcher`)

4. 📝 Subtitle (`bilibili_subtitle`)

5. ▶️ Player (`bilibili_player`)

6. 📤 Publisher (`bilibili_publisher`)