Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/rimagination/bili-note/bili-notenpx skills add Rimagination/bili-note --skill bili-notegit clone --depth 1 https://github.com/Rimagination/bili-noteWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/rimagination/bili-note/bili-note)<a href="https://agentmods.dev/skills/rimagination/bili-note/bili-note"><img src="https://agentmods.dev/badge/skills/rimagination/bili-note/bili-note.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00086 | $0.06324 |
| Opus 5 | $0.00043 | $0.03162 |
| Sonnet 5 | $0.00017 | $0.01265 |
| Haiku 4.5 | $0.00009 | $0.00632 |
Grade A, and why
bili-note scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl.exe -s http://localhost:3456/targets How it starts
The opening of the file, as written. The whole thing — 346 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bili Note
把 B站视频和图文动态变成可检索、可复用的 Markdown 知识笔记。视频优先拿字幕;字幕拿不到时再转写音频;图文优先抓正文、图片和代码块;用户要评论区时抓取评论并过滤无关讨论。
联网或登录态操作必须先使用 web-access。
网页 AI 字幕当前只支持 Chrome + web-access 路线:让已登录的 B站页面自己请求字幕接口。不要把 Edge、Playwright 临时浏览器或原生 CDP 端口当成等价替代,除非脚本已经明确支持。
Bili Note 与 DyNote 共享可复用本地资源。默认共享目录是 %USERPROFILE%\.cache\rimagination-notes,Qwen3-ASR 环境默认是 %USERPROFILE%\.cache\rimagination-notes\qwen3-asr-venv。如果任一 skill 已经安装过 Qwen3-ASR,另一个 skill 必须优先复用,不要重复安装。Hugging Face、Whisper 和 faster-whisper 缓存按本机通用缓存复用。
字幕密度与视觉理解
B 站视频优先拿字幕,但字幕/转写不等于完整理解。长视频如果字幕字数明显偏少,可能说明核心信息在画面、PPT、板书、代码演示、屏幕操作或无解说片段中。
- 归档后必须读取
metadata/note_budget.json。如果visual_dependency.needs_visual_review=true,不要把稀疏字幕写成完整学习笔记。 - 这种情况下,先采用更合适的技术:抽取代表性关键帧或截图,再做 OCR 或多模态视觉理解;必要时重新抓网页 AI 字幕或补本地自动语音识别。
- 如果当前接入的模型或工具不能看图,必须告诉用户:关键帧理解属于高级视觉能力,需要多模态模型、OCR 工具或人工查看;当前只能基于字幕/元数据生成有限笔记并标注覆盖范围。
- 最终笔记和用户回复要转述
visual_dependency.warnings,并说明是否已经补了视觉证据。
依赖与环境检查
默认路线尽量零第三方 Python 依赖:视频元数据、公开字幕、图文正文、图片清单、评论、归档、证据索引和笔记预算都用标准库完成。网页 AI 字幕和音频转写是增强路线,不是启动门槛。
第一次使用、换机器、用户怀疑依赖不全,或准备使用网页 AI 字幕 / 音频转写兜底时,先运行:
$skill = "$env:USERPROFILE\.codex\skills\bili-note"
$py = "python"
& $py "$skill\scripts\check_environment.py"
根据检查结果选择路线:
public_subtitles_comments_archive=OK:优先走默认字幕/图文、评论和归档流程。browser_ai_subtitles=OK:当公开接口只有ai-zh且subtitle_url为空时,走 Chrome +web-access网页 AI 字幕。audio_asr_fallback=OK:只有字幕和网页 AI 字幕都不可得、且用户确实需要完整转写时,才走音频转写。中文或未指定语言优先共享 Qwen3-ASR;明确外语视频优先 Whisper 系后端。- 某个增强能力缺失时,只说明该路线暂不可用;不要把它说成整个 skill 不可用。
- 网页登录态只通过
web-access已授权的 Chrome 页面使用;不要读取或复制 Cookie/profile,不要强制结束用户浏览器进程。没有 Chrome +web-access时就跳过网页 AI 字幕并说明覆盖范围。
默认流程
- 读清用户要什么:视频或图文链接、是否要评论区、保存路径、是否需要全文材料或只要提炼。
- 如果是首次使用、依赖状态不明、字幕抓取失败或用户要求音频转写,先用
check_environment.py判断当前可走路线。 - 优先用
run_bili_note.py一键完成可自动化部分。它会自动识别/video/BV...、/opus/...、/dynamic/...或纯 opus id。 - 如果是图文/动态,走
extract_bilibili_opus.py路线:抓正文、标题、作者、发布时间、图片、代码块、图文证据索引;用户加--comments时抓图文评论。 - 如果是视频,优先下载字幕:
- 普通字幕 URL 可用时,直接用
--download-subtitles。 - 如果普通接口显示
ai-zh但subtitle_url为空,不要说“没有字幕”;改走“网页 AI 字幕”流程。 - 如果字幕仍不可得,再按需要下载音频并用本地自动语音识别转写。中文优先 Qwen3-ASR,明确外语优先 Whisper 系后端。
- 普通字幕 URL 可用时,直接用
- 用户要求评论区时,用
--comments抓取主评论和子评论;写入笔记时过滤打卡、求资料、广告、闲聊等技术无关内容。 - 归档原始材料:把完整字幕或图文正文、图片、完整评论、元数据和 JSONL 索引存到知识库旁边的长期目录。
- 写前定标:必须先读取
metadata/note_budget.json,把推荐字数区间、压缩比目标、写作粒度、互动质量倍率、证据块数量和visual_dependency作为本次笔记的写作目标。视频按时长、字幕字数、证据块、评论量和互动质量定标;图文按正文长度、图片/代码/证据块、评论量和互动质量定标。- 如果
visual_dependency.risk是medium或high,先补关键帧/OCR/多模态视觉理解,或明确告诉用户当前缺少视觉理解能力,不能写成完整解析。
- 如果
- 按预算写 Markdown:默认写成“学习型笔记”,目标是让人或 Agent 像学完一节课或读完一篇教程一样获得概念、方法、判断标准、实践步骤和自测题;根据预算决定详略,不要把长课和短视频写成差不多字数。来源、覆盖范围和归档路径放到后半部分。正文证据默认用论文式编号
[1][2],不直接堆长证据 ID。 - 写后验收:用
score_bili_note.py校验笔记字数、压缩比、每分钟/每篇笔记密度和证据引用比例。评分只做 QA 和微调,不代替写前定标;太短时优先补“学习收获、知识地图、概念卡、实战流程、坑点、自测题”,不要只堆分P摘要或段落摘要。
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 346 lines · 86 tokens per session scan A c4d80165a0bb
bili-note is a skill published in the GitHub repository Rimagination/bili-note (281 stars, last pushed 2mo ago), licensed MIT. It adds 86 tokens to every session and 6,324 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
larksnap-fetch
把飞书/Lark 文档或普通网页抓取并保存到本地,也能编辑用户有权限的飞书文档,并用已登录浏览器执行一次网页搜索。用户要求下载、导出、抓取、写入飞书文档,或联网搜索资料/参考链接时使用本技能,即使没有提到 larksnap。底层通过技能自带 daemon 桥接已登录的 larksnap 浏览器扩展;arXiv 使用独立脚本。.
markdown-documents
Use when creating or editing Markdown documents, notes, reports, briefs, drafts, or other editable writing where Markdown should be the primary artifact format.
bilibili-to-doc
This skill should be used when the user asks to "提取B站视频", "B站视频转文档", "bilibili视频提取", "将B站视频转文字", or provides a bilibili.com video URL and wants to extract its content into a structured Markdown document. Automatically downloads AI subtitles (Chinese) via yt-dlp, parses SRT, and generates a well-formatted…
cv-update-review
Run a confirmation-gated CV update workflow for Word CV documents: set up a reusable profile, scan configured local activity sources plus public evidence such as PubMed through NCBI E-utilities and configured website/RSS feeds, prepare a CV update packet, ask the user to approve completed/public/accepted items, create…
install-by2kb
Install, configure, update, or diagnose by2kb for an Agent user. Defaults to local faster-whisper and local Markdown; use a cloud ASR provider only when the user explicitly chooses one.
pdf-ocr-to-markdown
OCR PDFs/images into user-facing Markdown with internal JSON/quality artifacts. Preserve originals; never create or modify PDFs. Use bundled scripts, not ad hoc OCR code. Mark visual pages pending unless images are actually inspected.