> 功能说明: 本技能涵盖 多种输入格式、专业的能力支持 等核心能力。

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：

Features

Installation

Requirements

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

依赖说明

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., config.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:

export BILIBILI_PERSIST=1

app = BilibiliAllInOne(persist=True)

When persistence is enabled:

• Credentials are auto-saved to .config.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()

⚠️ Security & Privacy

Credential Handling

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

Best Practices

1. 🧪 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).
2. 🔒 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.
3. 📁 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.
4. 🐳 Run in isolation — When possible, run this skill in an isolated container/environment and inspect network traffic.
5. 🌐 Verify network traffic — All HTTP requests go to Bilibili's official domains only. You can verify by monitoring outbound connections.
6. ❌ No exfiltration — This skill does NOT send credentials to any third-party service, analytics endpoint, or telemetry server.
7. 🔑 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

Credential Requirement by Module

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

Supported Categories

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

示例

python main.py hot_monitor get_hot '{"page_size": 10}'

python main.py hot_monitor get_trending '{"limit": 5}'

python main.py hot_monitor get_weekly

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

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.

Quality Options

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

Format Options

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

Examples

python main.py downloader get_info '{"url": "BV1xx411c7mD"}'

python main.py downloader get_formats '{"url": "BV1xx411c7mD"}'

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

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

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

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.

Supported Platforms

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

4. 📝 Subtitle (bilibili_subtitle)

5. ▶️ Player (bilibili_player)

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

Danmaku Modes

6. 📤 Publisher (bilibili_publisher)

Project Structure

> 详细代码示例已移至 references/detail.md

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

安装与配置

运行环境

• Agent平台: 支持SKILL.md的任意AI Agent( Code / Cursor / Codex / CLI等)
• 操作系统: Windows / macOS / Linux

第三方依赖

API Key 配置

• 本Skill基于Markdown指令,无需额外API Key(除内容中明确标注的外部API)

可用性分类

• 分类: MD+execute(纯Markdown指令,部分功能需要exec命令行执行能力)
• 说明: 基于Markdown的AI Skill,通过自然语言指令驱动Agent执行任务

使用场景

不适用于：需要人工判断的复杂决策场景

常见疑问

Q1: 如何开始使用Bilibili All In One？

A: 请先阅读使用流程章节，确认环境满足依赖说明中的要求。

Q2: 遇到错误怎么办？

A: 请参考错误处理章节，按照表格中的处理方式操作。

Q3: Bilibili All In One有什么限制？

A: 请参考已知限制章节了解具体限制。

功能边界

• 需要LLM支持，无LLM环境无法使用
• 复杂场景可能需要人工辅助判断
• 性能取决于底层模型能力

效能分析

优势对比

关键特性

• 自动化执行: B站全功能工具箱,热门监控等一体化运营
• 文件处理: 支持多种文件格式的读取、解析和写入操作
• API集成: 通过标准化接口调用外部服务并处理响应
• 命令执行: 在安全沙箱中执行系统命令并收集结果

快速入门指南

1. 配置API密钥: 在环境变量中设置对应的API Key
2. 初始化连接: 使用提供的凭证建立API连接
3. 调用接口: 传入必要参数执行API调用
4. 准备文件: 确认文件路径正确且格式受支持
5. 执行处理: 调用对应的处理函数
6. 查看结果: 检查输出文件或返回数据
7. 检查环境: 确认运行时和依赖已安装
8. 执行命令: 使用正确的参数格式执行
9. 查看输出: 检查命令输出和退出码

前置条件

• 已安装所需运行环境(参考依赖说明)
• 已获取必要的API密钥或访问凭证(如适用)
• 输入数据已准备就绪

异常应对机制

针对Bilibili All In One使用中可能遇到的常见问题,提供以下排查方案:

Bilibili All In One通用排查步骤

1. 检查输入参数: 确认所有必填参数已提供且格式正确
2. 查看日志输出: 定位具体错误行和异常类型
3. 验证环境配置: 确认依赖库版本和运行环境满足要求
4. 逐步调试: 缩小问题范围,隔离故障模块