Podcast Synthesis
Synthesize multiple podcast/interview transcripts from the same industry into a structured, multi-perspective analysis report. Accepts two input types: (A) 小宇宙 episode URLs (the skill will transcribe audio first, then synthesize), or (B) pre-existing transcript files (.txt, .docx, .md). Trigger when the user provides 2 or more 小宇宙 links, transcript files, or a mix — and asks for analysis, synthesis, a report, or even just "help me understand this industry". Also trigger when the user mentions 小宇宙, podcast transcripts, interview transcripts, or says things like "I have a few podcast conversatio
Install
npx skills add qingliu1617-art/xiaoyuzhou-synthesis-skill --skill podcast-synthesisgit clone --depth 1 https://github.com/qingliu1617-art/xiaoyuzhou-synthesis-skill.git /tmp/xiaoyuzhou-synthesis-skill
mkdir -p ~/.claude/skills
cp -r /tmp/xiaoyuzhou-synthesis-skill/skill ~/.claude/skills/podcast-synthesisPersonal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.
From the SKILL.md
# 播客/访谈综合分析 Skill ## 核心理念 单一访谈容易形成片面视角。这个skill的价值在于:把同一行业中不同背景、不同立场的人说的话放在一起,找到他们的共识、分歧和盲点,生成一份比任何单篇访谈都更有深度的行业洞察。 The core value: cross-referencing multiple voices reveals tensions, blind spots, and non-obvious truths that no single interviewee can see alone. --- ## 两种输入模式 | 模式 | 输入 | 处理 | |------|------|------| | **A:音频链接** | 小宇宙节目 URL | Step 0 → 转录为 .txt → Step 1–6 合成报告 | | **B:已有逐字稿** | .txt / .docx / .md 文件 | 直接进入 Step 1–6 | 两种模式可以混合使用(例如:2 个 URL + 1 个已有逐字稿文件)。 --- ## Step 0: 转录(仅限音频链接输入) **如果用户提供的是小宇宙节目链接**(URL 包含 xiaoyuzhou.fm),执行以下操作: ### 0.1 获取 API Key 在开始转录前,检查用户是否已提供千问 API Key: - 如果用户已在对话中提供 → 直接使用 - 如果未提供 → 询问:「请提供您的千问 API Key(阿里云 DashScope),用于音频转录。」 ### 0.2 运行转录脚本 使用 `scripts/transcribe.py` 处理每个 URL: ```bash python scripts/transcribe.py \ "https://www.xiaoyuzhou.fm/episode/xxx" \ "https://www.xiaoyuzhou.fm/episode/yyy" \ --api-key <QWEN_API_KEY> \ --output-dir <用户工作目录> ``` 脚本会自动: 1. 从小宇宙页面解析节目标题和音频下载地址(通过 `__NEXT_DATA__` JSON) 2. 下载 MP3 音频文件 3. 用 ffmpeg 将长于 25 分钟的音频切分为多段 4. 调用 `qwen-audio-turbo` API 逐段转录 5. 合并并保存为 `[节目标题].txt`,文件头包含来源 URL 和元数据 ### 0.3 逐字稿作为交付物 每个 `.txt` 逐字稿文件都是**独立交付物**,需保存到用户工作目录并展示给用户,即使他们没有明确要求。格式: ``` # 逐字稿:[节目标题] **播客**:[播客名称] **发布日期**:[日期] **来源**:[小宇宙 URL] **转录模型**:qwen-audio-turbo --- [逐字稿正文...] ``` ### 0.4 错误处理 | 错误类型 | 处理方式 | |----------|----------| | 页面无法访问 / 结构变化 | 告知用户,跳过该 URL,继续处理其余 | | API Key 无效 | 停止,提示用户检查 API Key | | 音频下载超时 | 重试 1 次,失败则告知用户 | | 转录 API 失败 | 自动重试 3 次(指数退避),最终失败则保留已转录部分并标注 | ---
Details
- Skill name
podcast-synthesis- Repository
- qingliu1617-art/xiaoyuzhou-synthesis-skill
- Path
skill/SKILL.md- Repo updated
- 26 March 2026
FAQ
What does the Podcast Synthesis skill do?
Synthesize multiple podcast/interview transcripts from the same industry into a structured, multi-perspective analysis report. Accepts two input types: (A) 小宇宙 episode URLs (the skill will transcribe audio first, then synthesize), or (B) pre-existing transcript files (.txt, .docx, .md). Trigger when the user provides 2 or more 小宇宙 links, transcript files, or a mix — and asks for analysis, synthesis, a report, or even just "help me understand this industry". Also trigger when the user mentions 小宇宙, podcast transcripts, interview transcripts, or says things like "I have a few podcast conversatio
How do I install the Podcast Synthesis skill?
skills CLI (Claude Code, Cursor, Codex and more): npx skills add qingliu1617-art/xiaoyuzhou-synthesis-skill --skill podcast-synthesis. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.
Is Podcast Synthesis popular?
It lives in qingliu1617-art/xiaoyuzhou-synthesis-skill, which has 6 GitHub stars and 1 forks.
Related skills
Clarity Mapping
9duvoai/skills
Map how a process works with Duvo Clarity. Use when the user asks to map, document or understand how their team does a piece of work, to interview people about a process, or to analyze what those interviews found. Sets up a Duvo workspace if the user has none, creates the process, invites people to short voice interviews within the plan's allowance, tracks who has done theirs, and analyzes the process map.
Music Generation
8Cam10001110101/claude-skills-base
Tools, patterns, and utilities for generating professional music with realistic instrument sounds. Write custom compositions using music21 or learn from existing MIDI files.
Image to Code
5catch-the-wave/fullstack-ios-claude-skills
Elite website image-to-code skill for Codex. For visually important web tasks, it must first generate the design image(s) itself, deeply analyze them, then implement the website to match them as closely as possible. In Codex, it must prefer large, readable, section-specific images instead of tiny compressed boards, generate fresh standalone images for sections or detail views instead of cropping old ones, avoid lazy under-generation, avoid cards-inside-cards-inside-cards UI, and keep the hero clean, spacious, readable, and visible on a small laptop.
Diagram Tour
6centraltowerlabs/diagram-tour
Generate a narrated architecture-tour video from a codebase. End-to-end: codebase analysis → Graphviz .dot → narration script → narrated MP4. Three confirmation gates between user and final video. Use when the user asks to "explain this codebase", "tour the architecture", "make a video about how X works", or invokes /diagram-tour explicitly. Output: a versioned MP4 at <project>/.diagram-tour/renders/. Expensive — confirm runtime expectations (~15-30 min on modern hardware, longer on lighter machines) before kicking off.
Kurzgesagt Style Explainer
5maxeffortyt/kurzgesagt-style-explainer
Make a short flat-vector science explainer (about 60 s) in the measured visual style of Kurzgesagt – In a Nutshell, style only (no characters, logos or footage), with Remotion, an ElevenLabs voice-over synced by word timestamps, and a measurement loop that scores the finished render against the style. Use when asked for a Kurzgesagt-style or "flat vector space-y" animated explainer, or when you need to measure a video style in numbers and hit those numbers.
Code to Music
8Cam10001110101/claude-skills-base
Tools, patterns, and utilities for creating music with code. Output as a .mp3 file with realistic instrument sounds. Write custom compositions to bring creativity to life through music. This skill should be used whenever the user asks for music to be created. Never use this skill for replicating songs, beats, riffs, or other sensitive works. The skill is not suitable for vocal/lyrical music, audio mixing/mastering (reverb, EQ, compression), real-time MIDI playback, or professional studio recording quality.
Browse the most popular Claude skills, all images, video & audio skills, or write your own with the skills generator.