SkillsAdd

smithery/haizhouyuan

bilibili-up-batch

批量下载 B 站 UP 最近 N 条视频到 imports，并用 video_pipeline 生成证据?

Installation

$ npx skills add smithery/haizhouyuan --skill bilibili-up-batch
View on skills.sh ↗ Website ↗

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

bilibili-subtitle Similar name
hamsterider-m/bilibili-subtitle

从 Bilibili 视频提取字幕、转录无字幕视频、生成结构化摘要。触发条件：Bilibili URL (bilibili.com)…

1.4K installs
bilibili-downloader Similar name
serpdownloaders/skills

Download Bilibili videos

1.1K installs
bilibili-analyzer Similar name
aidotnet/moyucode

自动分析B站视频内容，下载视频并拆解成帧图片，使用AI分析并生成详细的专题文档或实操教程。

1.1K installs
bilibili-cli Similar name
public-clis/bilibili-cli

CLI skill for Bilibili (哔哩哔哩, B站) with token-efficient YAML output for AI agents to browse…

686 installs
bilibili-downloader Similar name
958877748/skills

Download Bilibili videos. Extracts video and audio streams separately.

560 installs
bilibili-render-pdf Similar name
wdkns/wdkns-skills

Generate a professional, detailed, figure-rich LaTeX course note and final PDF from a Bilibili …

352 installs

Also in this package

Other skills from smithery/haizhouyuan.

npx skills add smithery/haizhouyuan
topic-research 0 robot-update-package 0 capture-idea 0

Browse all from smithery/haizhouyuan

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Skill metadata

Parsed from SKILL.md frontmatter.

Allowed toolsvideo_pipeline, workspace, shell, python

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 2,091 B
  • docs SUMMARY.md 154 B

History

  1. First recorded snapshot · 0 installs

SKILL.md

触发条件

当用户给出一个 B 站 UP 主页（space.bilibili.com/<mid> 或 b23.tv 短链），并要求“拉取最近 N 条视频并批处理”为研究资产时使用。

目标

  1. 下载：把最新 N 条视频下载到 imports/content/videos/bilibili/<mid>/
  2. 分析：对每个视频运行 videopipeline，产出 state/video-analyses/<analysisid>/evidence.json 与 evidence_compact.md
  3. （可选）后续由 digest-content 读取 evidence_compact.md 生成 digest 并归档到 topic

依赖

  • ffmpeg（系统已安装即可）
  • yt-dlp（建议安装到仓库 venv：.venv/bin/python -m pip install yt-dlp）

硬约束

  • 不在 git 里提交下载的视频文件（imports/ 默认不入 git）
  • video_pipeline 只处理本地文件；下载是独立步骤
  • 不写入任何 cookie/token 到仓库（如需登录态下载，只允许通过环境变量/本地配置注入）

SOP（推荐用 worker panes）

  1. Controller pane（主控）先设置接收通知

- 运行：scripts/tmuxsetcontroller_pane.sh

  1. Worker pane：批量下载 + 分析

- 对单个 UP（最近 30 条）： - python3 scripts/bilibiliupbatch.py --up '<spaceorb23url>' --limit 30 --download --analyze --enable-ocr - 若你要节省时间（先跑转写，稍后再 OCR）： - python3 scripts/bilibiliupbatch.py --up '<spaceorb23url>' --limit 30 --download --analyze

  1. Worker 完成后通知 controller

- scripts/tmuxnotifycontrollerdone.sh --topic <topicid或占位> --record <state/runs/...json> --status done

  1. 生成 digest（可选但推荐）

- 对每个 state/video-analyses/<analysisid>/evidencecompact.md 使用 digest-content 生成 digest； - 如需归档到 topic：再用 topic-ingest 更新 sources.md/timeline.md。