Agent skill

Wjs Mining Voicedrop

by jianshuo in jianshuo/claude-skills

A skill your agent uses when 王建硕 wants to turn his uploaded VoiceDrop voice memos into 微信公众号 article drafts — pulling the unprocessed recordings sitting on jianshuo.dev/files (the R2 inbox)…

MITAuto-check: notesMedia & Creative

Install Wjs Mining Voicedrop

skills CLI
$ npx skills add jianshuo/claude-skills --skill wjs-mining-voicedrop -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install jianshuo/claude-skills wjs-mining-voicedrop --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/jianshuo/claude-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/wjs-mining-voicedrop .claude/skills/wjs-mining-voicedrop && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
wjs-mining-voicedrop
GitHub stars
131
Token cost
~1.3k tokens
SKILL.md length
302 words
Files
4 (incl. scripts)
Skills in repo
38
Repo updated
First seen
Licence
MIT

At a glance

A skill your agent uses when 王建硕 wants to turn his uploaded VoiceDrop voice memos into 微信公众号 article drafts — pulling the unprocessed recordings sitting on jianshuo.dev/files (the R2 inbox)…

  • Works in 4 steps: 定位脚本 + 载入环境(不依赖当前目录) → 列收件箱 → 逐条闭环(串行,一条跑完再下一条) → …
  • Transcribing them
  • SKILL.md covers Core Principle, When This Skill Fires, When NOT to use and 前置, plus 4 more sections
  • Runs Shell scripts from its folder; calls ffprobe; needs FILES_TOKEN

What it does

Wjs Mining Voicedrop is an agent skill from jianshuo/claude-skills. Use when 王建硕 wants to turn his uploaded VoiceDrop voice memos into 微信公众号 article drafts — pulling the unprocessed recordings sitting on jianshuo.dev/files (the R2 inbox), transcribing them, and mining articles from each. Triggers — "处理 VoiceDrop 录音", "把新录音挖成文章", "口述备忘变文章", "处理一下我的录音", "/wjs-mining-voicedrop".

Its SKILL.md is about 1.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 5 other files, including scripts (for example `agents/interface.yaml` and `scripts/voicedrop-inbox.sh`).

It sits in Media & Creative, covering Transcription, Email management and Blog and article writing. The repository describes itself as: 13 Claude Code skills for video production (transcribe / translate / dub / multicam / subtitles / reframe) + WeChat publishing. Compatible with Claude Code, OpenAI Codex CLI… The licence is MIT.

When your agent uses it

  • Transcribing them
  • Mining articles from each
  • — 处理 VoiceDrop 录音
  • /wjs-mining-voicedrop

Example prompts

  • “处理 VoiceDrop 录音”
  • “把新录音挖成文章”
  • “口述备忘变文章”
  • “/wjs-mining-voicedrop”

Requirements

  • A Bash shell
  • A credential in FILES_TOKEN

Workflow steps

4 steps, taken from the step headings in SKILL.md.

  1. 定位脚本 + 载入环境(不依赖当前目录)
  2. 列收件箱
  3. 逐条闭环(串行,一条跑完再下一条)
  4. 汇报

What it can do on your machine

Read from SKILL.md and the folder at commit b2690f5. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 1 file in scripts/ (Shell), which the agent can run.

    Shell commands in SKILL.md call:

    • ffprobe

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names these keys or tokens, usually read from environment variables:

    • FILES_TOKEN

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Wjs Mining Voicedrop loads about 1.3k tokens when it runs. Until then it costs about 83 tokens; SKILL.md has 302 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~83
When it runs · the whole SKILL.md, loaded when a task matches
~1.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

  • NoteMentions a .env fileSKILL.md:31
    - `~/code/.env` 里有 `FILES_TOKEN`(收件箱鉴权)和火山 ASR creds(`VOLC_ASR_*` / `VOLC_TTS_*`,转写用)。`set -a; source ~/code/.env; set +
  • NoteMentions a .env fileSKILL.md:35
    ark-empty` / `delete`,token 运行时从 `~/code/.env` 读,绝不落代码)。`list` 只列未处理;`mark-done`/`mark-empty` 写处理标记;`delete` 只给手动清理用,成文流
  • NoteMentions a .env fileSKILL.md:41
    set -a; source ~/code/.env; set +a    # FILES_TOKEN + 火山 ASR creds
  • NoteMentions a .env fileSKILL.md:52
    失效)→ 报「收件箱连不上或 FILES_TOKEN 失效,检查 `~/code/.env`」并停,**不进入循环**。
  • NoteMentions a .env fileSKILL.md:100
    | `~/code/.env` | `FILES_TOKEN` + 火山 ASR creds |

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from jianshuo/claude-skills at commit b2690f5, republished under its MIT licence (© jianshuo). 302 words, ~1,276 tokens.

Download SKILL.mdSave it as .claude/skills/wjs-mining-voicedrop/SKILL.md (or your agent's skills folder). This skill also uses 3 other files; get the full folder from GitHub.
name
wjs-mining-voicedrop
description
Use when 王建硕 wants to turn his uploaded VoiceDrop voice memos into 微信公众号 article drafts — pulling the unprocessed recordings sitting on jianshuo.dev/files (the R2 inbox), transcribing them, and mining articles from each. Triggers — "处理 VoiceDrop 录音", "把新录音挖成文章", "口述备忘变文章", "处理一下我的录音", "/wjs-mining-voicedrop".

wjs-mining-voicedrop

VoiceDrop 收件箱(jianshuo.dev/files 上的 VoiceDrop-*.m4a)→ 逐条转写 → 交给 wjs-mining-articles 出公众号草稿。这是 VoiceDrop iOS app(开口即录、停即上传)的 Mac 端闭环。

本 skill 自身的产出 = ① 公众号草稿(~/code/wechat-publish/)+ ② 本地音频/SRT 存档(~/code/voicedrop/archive/)+ ③ R2 上的处理标记(articles/<stem>.json 或 .empty)+ ④ 一份批次报告(处理几条、各出几篇、哪些标了无语音及原因、还剩几条未处理)。 完整接口契约见 agents/interface.yaml。

Core Principle

复用,不重写。 本 skill 只做两件本身没有的事:收件箱的进出(列/下载/标记)和逐条编排。转写交 wjs-transcribing-audio,成文交 wjs-mining-articles,一行都不重写。

R2 永不删,用标记文件表示处理状态。 音频一直留在 R2,直到用户自己在 app 里删。「未处理」= 还没有 articles/<stem>.json(已成文)也没有 articles/<stem>.empty(无语音)标记的 VoiceDrop-*.m4a;list 已自动只列未处理的。一条成功成文后写 mark-done,没语音/损坏写 mark-empty——两者都让这条不再被重复处理。绝不 delete(delete 只留给用户在 app 里手动清理)。

When This Skill Fires

  • 用户说「处理 VoiceDrop 录音」「把新录音挖成文章」「处理一下我的口述」
  • 用户跑 /wjs-mining-voicedrop

When NOT to use

  • 已经有 SRT → 直接 wjs-mining-articles
  • 音频不在 R2 收件箱(本地散文件)→ 直接 wjs-transcribing-audio 出 SRT,再 wjs-mining-articles
  • 桶里是别的机器传的非录音文件 → 本 skill 只认 VoiceDrop-*.m4a 前缀,其余不碰

前置

  • ~/code/.env 里有 FILES_TOKEN(收件箱鉴权)和火山 ASR creds(VOLC_ASR_* / VOLC_TTS_*,转写用)。set -a; source ~/code/.env; set +a。

Workflow

唯一的新增代码:scripts/voicedrop-inbox.sh(list / download / mark-done / mark-empty / delete,token 运行时从 ~/code/.env 读,绝不落代码)。list 只列未处理;mark-done/mark-empty 写处理标记;delete 只给手动清理用,成文流程不调它。

Step 0 · 定位脚本 + 载入环境(不依赖当前目录)
bash
INBOX=~/.claude/skills/wjs-mining-voicedrop/scripts/voicedrop-inbox.sh
set -a; source ~/code/.env; set +a    # FILES_TOKEN + 火山 ASR creds

用绝对路径 $INBOX 调脚本——不要写成 scripts/voicedrop-inbox.sh,那依赖「人恰好在 skill 根目录」这个隐藏假设,换目录就崩。

Step 1 · 列收件箱
bash
"$INBOX" list      # 打印未处理的 VoiceDrop-*.m4a,一行一个
  • 命令非零退出(网络不通 / token 失效)→ 报「收件箱连不上或 FILES_TOKEN 失效,检查 ~/code/.env」并停,不进入循环。
  • 输出为空 → 报「收件箱没有新录音」结束。
  • 非空 → 拿到这一批文件名。
Step 2 · 逐条闭环(串行,一条跑完再下一条)

串行。批次韧性:单条任何一步失败 → 记录原因、跳到下一条、绝不中止整批、绝不漏标。 每条录音最终必须落到三个终态之一:已成文(mark-done)/ 无语音(mark-empty)/ 失败(不标,留待下次)——绝不「处理了却什么都没标」。对每个 <name>:

  1. 下载存档:
    bash
    "$INBOX" download <name> ~/code/voicedrop/archive
    音频落 ~/code/voicedrop/archive/<name>。R2 上的原件始终保留,本地这份只是离线副本。
  2. 快速看一眼是不是真录音:
    bash
    dur=$(ffprobe -v error -show_entries format=duration -of csv=p=0 "$audio" 2>/dev/null)
    dur 为空(非音频/损坏)→ "$INBOX" mark-empty <name> corrupt;< 1.0 秒(误传/静音)→ "$INBOX" mark-empty <name> silent。标完报告用户,跳到下一条——这条已是终态,不再重复处理。
  3. 转写 → SRT:载入 wjs-transcribing-audio(中文走火山豆包 volc_asr_stream.py + build_srt_from_asr.py + 在 session 内做 AI 润色改错别字)。SRT 落 ~/code/voicedrop/archive/<stem>.srt。
    • 转写出来是空的(有声音但 ASR 一字未出,多是环境音)→ "$INBOX" mark-empty <name> no-speech,报告、跳下一条。
  4. 挖文章:把这个 SRT 交给 wjs-mining-articles 跑它的完整流程——出选题清单(它的人工闸,照走别跳)、成文、建微信草稿。语音备忘多是短独白单主题,清单常只有 1 条,照常让用户确认。
  5. 成文成功后写处理标记(出了至少一篇草稿、用户没中止):把这一条挖出的文章拼成 v2 JSON({"schema":2,"status":"ready","sourceAudio":"<name>","articles":[{"title","body"},…]},可含 transcript/srt)写到临时文件,再:
    bash
    "$INBOX" mark-done <name> /tmp/<stem>.json
    这条就标成已成文、app 里也能看到,且不会被服务器或下次再挖。 转写失败 / 用户没勾任何选题 / 挖不出文章 → 不标 done 也不标 empty,留未处理,下次再来,报告原因。
Step 3 · 汇报

处理了几条、各挖出几篇草稿(落在 ~/code/wechat-publish/)、哪些标了无语音及原因(corrupt/silent/no-speech)、本地存档路径、R2 还剩几条未处理。

标记安全红线

download(存档) → 判别 → 成文 ? mark-done : 无语音 ? mark-empty : 留着不标
                                                     ↑ 绝不 delete
  • 绝不 delete。 音频永远留在 R2,删除只属于用户在 app 里的手动操作。
  • 每条都有终态。 成文 → mark-done;损坏/静音/无语音 → mark-empty(带 reason);真失败(转写报错、用户中止、没挖出文章)→ 不标,留未处理下次再试。绝不出现「跑过一遍却没留任何标记」——那会让这条每次都被重新处理。

复用边界

复用用法
wjs-transcribing-audio每条音频 → SRT(中文火山豆包,含润色改错别字)
wjs-mining-articles每个 SRT → 选题清单 → 成文 → 微信草稿(含它自己的人工闸)
~/code/.envFILES_TOKEN + 火山 ASR creds
VoiceDrop app上游:文件名形如 VoiceDrop-<时间戳>-<时长>-<星期>-<时段>[-<城市-城区>].m4a(全 ASCII)。本 skill 靠 VoiceDrop- 前缀 + .m4a 后缀认领;中间的时长/星期/时段/地点是上下文,成文时可借来判断这条录音是何时何地的口述
服务器 miner(~/code/voicedrop/mining/mine.py,每 2h)同一套标记约定:成文写 articles/<stem>.json、无语音写 articles/<stem>.empty、永不删音频。它会自动处理收件箱,所以本 skill 跑时 list 常常已经空了——这是预期,本 skill 是手动补位

本 skill 唯一新增代码:scripts/voicedrop-inbox.sh。

Common Mistakes

  • delete 音频 —— 红线。成文流程永不 delete,只 mark-done/mark-empty;delete 仅用户在 app 里手动用。
  • 跑过一条却不标记 —— 无语音/损坏的也要 mark-empty,否则它每次都被重新下载转写,永远「待处理」。
  • 转写失败/用户没勾选也硬标 done —— 真失败就留未处理(不标),下次再试;只有出了草稿才 mark-done。
  • 跳过 wjs-mining-articles 的选题闸自己硬写 —— 那个闸是它的设计,照走。
  • 把桶里非 VoiceDrop 文件也当源 —— 只认 VoiceDrop-*.m4a 前缀;list 也只列未处理的。
  • 误传/0 秒/环境音当真录音反复试 —— 先 ffprobe 看时长,空/损坏/<1s 直接 mark-empty,别送去转写。

© jianshuo, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 3 other files (scripts) in wjs-mining-voicedrop of jianshuo/claude-skills.

  • SKILL.md
  • SKILL.md.bak.20260620-095943
  • agents/interface.yaml
  • scripts/voicedrop-inbox.sh

Open the folder on GitHubat commit b2690f5

Compare with similar skills

Wjs Mining Voicedrop next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Wjs Mining Voicedrop compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Wjs Mining Voicedrop this skilljianshuo/claude-skills131—~1.3kAutomated safety check: NotesMIT
Transcribegnekt/My-Brain-Is-Full-Crew3.9k—~5kAutomated safety check: PassCustom licence
Voice Personadavepoon/buildwithclaude3.6k—~947Automated safety check: PassMIT
Medium Writingericrisco/rsc-harness180—~3.3kAutomated safety check: PassMIT
Call Transcript Reliability GateCALLE-AI/awesome-phone-call-agents107—~1.3kAutomated safety check: PassMIT
HyperFrames Media Useheygen-com/hyperframes60k—~2.4kAutomated safety check: PassApache-2.0

Similar skills

  • Transcribe

    gnekt/My-Brain-Is-Full-Crew

    Process audio recordings, meeting transcripts, podcasts, or lectures.

    3.9k GitHub stars~5k tokensUpdated 3 mo ago
    Media & CreativeAuto-check passed
  • Voice Persona

    davepoon/buildwithclaude

    让 Agent 变成能语音对话的机器人:语音文件转文字(支持微信 silk 格式)+ 多音色人格回复(Edge TTS 免费中文音色),全本地零 API 成本。Voice persona chat: transcribe voice messages (incl.

    3.6k GitHub stars~947 tokensUpdated 2 days ago
    Media & CreativeAuto-check passed
  • Medium Writing

    ericrisco/rsc-harness

    A skill your agent uses when crafting or rewriting ONE Medium story so it earns reads, claps and Boost curation — the title, subtitle and kicker as a hook system, fixing a story with views but a low…

    180 GitHub stars~3.3k tokensUpdated yesterday
    Writing & ContentAuto-check passed
  • Call Transcript Reliability Gate

    CALLE-AI/awesome-phone-call-agents

    Offline experimental CALL-E transcript auditor that grades a returned transcript RELIABLE, SUSPECT, or UNUSABLE from text-visible ASR-hallucination symptoms before anything acts on it, and crafts…

    107 GitHub stars~1.3k tokensUpdated yesterday
    Media & CreativeAuto-check passed
  • HyperFrames Media Use

    heygen-com/hyperframes

    Finds, generates and edits media for HyperFrames video projects: music, sound effects, images, icons, logos, voiceovers, captions and color grades.

    60k GitHub stars~2.4k tokensUpdated today
    Media & CreativeAuto-check passed
  • Native Subtitle Quote Image

    chengyi-ai/native-subtitle-quote-image

    将本地视频或用户有权处理的在线视频,经过来源获取、文字稿定位、选题选句、精确取帧、紧凑裁切、拼图和逐张质检,制作成 3:4 或保留画面原比例的视频字幕长图。支持两种明确分开的输出:保留画面内已烧录字幕的原生字幕模式,以及把已审核的时间点与台词绘制到真实视频帧上的脚本字幕模式。用户要求原生字幕截图、字幕帧拼图、YouTube…

    2.6k GitHub stars~2.4k tokensUpdated today
    Media & CreativeAuto-check passed

More from jianshuo/claude-skills

All 38 skills in this repo
  • Wjs Segmenting Video

    jianshuo/claude-skills

    A skill your agent uses when the user has a long-form video (interview / lecture / podcast / conversation) and a transcript SRT, and wants to extract 3–6 stand-alone topical short clips from it.

    131 GitHub stars~3.5k tokensUpdated 1 mo ago
    Auto-check passed
  • Wjs Uploading Video

    jianshuo/claude-skills

    Upload one or many videos to YouTube. An agent skill from jianshuo/claude-skills.

    131 GitHub stars~2.1k tokensUpdated 1 mo ago
    Auto-check passed
  • Wjs Converting Wp To Hugo

    jianshuo/claude-skills

    A skill your agent uses when migrating a WordPress site to a Hugo static site on GitHub Pages from a WXR export (.xml) plus the wp-content/uploads folder — preserving /archives/<id/ URLs, localizing…

    131 GitHub stars~1.8k tokensUpdated 1 mo ago
    Auto-check passed
  • Wjs Burning Subtitles

    jianshuo/claude-skills

    A skill your agent uses when the user has a video + an SRT and wants the subtitles either burned into the pixels (libass, always-visible) or soft-muxed as a togglable track.

    131 GitHub stars~3k tokensUpdated 1 mo ago
    Auto-check passed
  • Wjs Cleaning Spam

    jianshuo/claude-skills

    A skill your agent uses when the user complains about spam on his X/Twitter posts — 同城面付 / 寻固炮 / 线下上门 / 免费破处 这类引流号在他推文下刷的 emoji 垃圾回复 — and wants them removed.

    131 GitHub stars~532 tokensUpdated 1 mo ago
    Auto-check passed
  • Wjs Creating Video Book

    jianshuo/claude-skills

    A skill your agent uses when the user wants a book turned into YouTube chapter videos — 每章用 VoiceDrop 读书的有声书 mp3 做音轨,配 GPT Image 2 画面和中心思想大字,输出 1920×1080 横屏视频发 YouTube。Triggers — "把这本书做成视频"…

    131 GitHub stars~1.8k tokensUpdated 1 mo ago
    Auto-check passed

Questions about Wjs Mining Voicedrop

What does Wjs Mining Voicedrop do?

A skill your agent uses when 王建硕 wants to turn his uploaded VoiceDrop voice memos into 微信公众号 article drafts — pulling the unprocessed recordings sitting on jianshuo.dev/files (the R2 inbox)…. Wjs Mining Voicedrop is an agent skill from jianshuo/claude-skills.dev/files (the R2 inbox), transcribing them, and mining articles from each.

When should I use Wjs Mining Voicedrop?

Wjs Mining Voicedrop fits situations like: transcribing them; mining articles from each; — 处理 VoiceDrop 录音; /wjs-mining-voicedrop.

How do I install Wjs Mining Voicedrop in Claude Code?

Run `npx skills add jianshuo/claude-skills --skill wjs-mining-voicedrop -a claude-code`. Or copy the skill folder (wjs-mining-voicedrop in jianshuo/claude-skills) into .claude/skills/wjs-mining-voicedrop in your project. Claude Code loads it when a task matches its description.

How do I install Wjs Mining Voicedrop in Codex?

Run `npx skills add jianshuo/claude-skills --skill wjs-mining-voicedrop -a codex`. Or copy the skill folder (wjs-mining-voicedrop in jianshuo/claude-skills) into .agents/skills/wjs-mining-voicedrop in your project. Codex loads it when a task matches its description.

Can I use Wjs Mining Voicedrop in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add jianshuo/claude-skills --skill wjs-mining-voicedrop -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/wjs-mining-voicedrop, .gemini/skills/wjs-mining-voicedrop, .github/skills/wjs-mining-voicedrop and .opencode/skills/wjs-mining-voicedrop in your project.

What does Wjs Mining Voicedrop need to run?

Going by SKILL.md and its folder, Wjs Mining Voicedrop needs a shell for the scripts in its folder, the command-line tools its instructions call (ffprobe) and credentials named FILES_TOKEN. Our summary lists: A Bash shell; A credential in FILES_TOKEN.

Does Wjs Mining Voicedrop access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Wjs Mining Voicedrop safe to install?

Our automated static check of SKILL.md found notes only (mentions a .env file), nothing it rates as a warning. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Wjs Mining Voicedrop use?

Wjs Mining Voicedrop is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Wjs Mining Voicedrop use?

About 1.3k tokens (SKILL.md is roughly 5.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Wjs Mining Voicedrop?

Skills that share tags, products or a category with Wjs Mining Voicedrop: Transcribe (gnekt/My-Brain-Is-Full-Crew, 3.9k stars), Voice Persona (davepoon/buildwithclaude, 3.6k stars), Medium Writing (ericrisco/rsc-harness, 180 stars) and Call Transcript Reliability Gate (CALLE-AI/awesome-phone-call-agents, 107 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Wjs Mining Voicedrop?

jianshuo (a GitHub user) maintains it in jianshuo/claude-skills, which has 131 GitHub stars. The repository holds 38 skills in this directory. The repository was last updated on August 20, 2026.

Source: jianshuo/claude-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.