Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Podcast Generator · staruhub | DecimalAI

design mediaby staruhub

Podcast Generator

staruhub/podcast-generator

Source staruhub/claudeskills ↗License MIT

用火山引擎 Podcast AI 模型生成中文双人对话播客。当用户要把文章、报告、话题文本转成播客音频、生成对话式音频内容时使用，需要环境具备火山引擎 APP_ID 和 ACCESS_KEY。支持 mp3/ogg_opus/pcm/aac、语速调节、自定义音色、断点续传。不用于：单人朗读式 TTS（用普通语音合成）、英文播客（模型主要优化中文）、播客文稿本身的撰写（先用写作类 skill 产出文本再来）。

SkillScore82Maintained

Automatic

Marketing & Content pack

Last verified 8/18/2026

What it changes

Passed 18 of 22 test cases with the skill, and 8 of 22 without it.

Without skill36%

With skill82%

Checks it passed that the model alone missed

Points to references/api_reference.md
Identifies 24000 Hz as the default sample rate
Identifies websockets as the required Python package

and 8 more in the full report

10 fixed8 passed either way4 still failing0 broke

Pass-rate lift: +45 pp
Tokens: -13%
Agent turns: 0%
Response time: -56%

Model gemini-3.6-flashMeasured 8/18/2026v118 of 22 cases passed with skill

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/staruhub-podcast-generator/SKILL.md https://app.decimal.ai/s/staruhub-podcast-generator/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add staruhub-podcast-generator, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/staruhub-podcast-generator/ and pins it in .decimal/skills.lock

decimalai skills add staruhub-podcast-generator

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/staruhub-podcast-generator/SKILL.md https://app.decimal.ai/s/staruhub-podcast-generator/SKILL.md

Raw file: /s/staruhub-podcast-generator/SKILL.md ↗ · machine-readable summary at /s/staruhub-podcast-generator.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 10 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 10

Prompt

这里有一份中文编程教学讲义，我希望把它转换成两人对谈风格的播客音频方便学生听讲。因为是教学内容，希望语速稍微放慢一点，讲得更清晰一些。请帮忙跑一下生成流程，最后回报生成的音频路径和文件大小。

What was checked, without and with the skill
Checked	Without skill	With skill
Sets speech_rate parameter to -20 for instructional content	Missed	Passed
Uses scripts/generate_podcast.py to run the audio generation task	Missed	Passed

case-02 · gemini-3.6-flash · 8/18/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md63 lines3 filesincludes scripts

播客生成器（火山引擎双人对话）

把中文文本变成双人对话播客音频，接口封装在 scripts/generate_podcast.py。

验收标准（每次生成完成前自查）

音频文件已落盘，把实际路径与文件大小回报用户
输入文本 ≤25,000 字符（超长必须先与用户确认拆分方案，不能让模型静默截断）
生成中断时使用了 retry_info 续传而不是从头重来
参数选择有依据：分发用 mp3，后期加工用 pcm；教学内容语速 -20 左右
凭证缺失时未硬试：告知用户去火山引擎控制台（speech/service/10028）获取

不做什么

不写播客文稿——输入文本的质量是上游任务（文稿创作找写作类 skill）
不做单人朗读、配音、音效制作
不在输出里回显用户的 ACCESS_KEY

工作流程

1. 准备输入

必需：中文文本（≤25k 字符）+ APP_ID + ACCESS_KEY（无则告知获取方式后停止）。可选：格式（默认 mp3）/ 采样率（默认 24000）/ 语速（-50~100，0=正常，100=2 倍速）/ 音色 / 开场音乐（默认关）。

最佳文本长度 500-3000 字——播客时长与听感的最优区间；一篇长文建议先摘要再生成。

2. 生成

python scripts/generate_podcast.py \
  --text "播客话题或内容文本" \
  --output "/path/to/output.mp3" \
  --app-id "$VOLC_APP_ID" --access-key "$VOLC_ACCESS_KEY" \
  --format mp3 --sample-rate 24000 --speech-rate 0

脚本会流式接收音频、按轮次显示进度、落盘后返回统计（大小/轮次数）。 Python 模块调用、自定义音色 ID、断点续传 retry_info 的写法见脚本内 docstring 与 references/api_reference.md。

3. 交付

Track record

Benchmark

18 / 22

test cases passed, vs 8 / 22 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

staruhub / claudeskills

lab/Geek-skills-podcast-generator/SKILL.md

View on GitHub ↗

License MITImported 8/3/2026Version v1Bundled 2 files · scripts

More skills from the same source and nearby in the registry, with their headline benchmark lift.