Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
Mae avatar

Gencast

  • 1 repo stars
  • Updated July 25, 2026
  • cadrianmae/claude-marketplace

Auto-generate podcasts from documents with gencast (always --minimal), supporting planning, styles, audiences, voices, and dialogue/plan saving.

About

Gencast auto-invokes the gencast tool to turn documents into podcasts, always using the --minimal flag. It supports the full range of CLI options including planning, styles, audiences, voices, and saving the dialogue or plan. This makes it a quick path from written material to conversational or lecture-style audio.

  • Generates podcasts from documents
  • Always uses --minimal flag
  • Planning, styles, audiences, voices options
  • Saves dialogue and plan

Gencast by the numbers

  • Data as of Jul 26, 2026 (Skillselion catalog sync)
/plugin marketplace add cadrianmae/claude-marketplace
/plugin install gencast@cadrianmae-claude-marketplace

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
repo stars1
Last updatedJuly 25, 2026
Repositorycadrianmae/claude-marketplace

What it does

Auto-generate podcasts from documents with gencast (always --minimal), supporting planning, styles, audiences, voices, and dialogue/plan saving.

README.md

Version License: MIT

gencast Plugin for Claude Code

Auto-invoke gencast to generate conversational podcasts from documents. Always uses --minimal flag for clean context management.

Features

  • Auto-invocation - Claude automatically suggests podcast generation when you mention "podcast", "audio", or "dialogue"
  • Full CLI support - All gencast options available: styles, audiences, voices, planning
  • Minimal output - Always uses --minimal flag to reduce context bloat
  • Smart workflows - Validates inputs, checks installation, provides clear progress updates
  • Reference docs - Voice options and style/audience combinations documented

Installation

The gencast CLI tool is required:

pip install gencast

Quick Start

Auto-invocation

Just mention podcast generation in your conversation:

"Can you convert this lecture.md to a podcast?"

Claude will automatically use gencast with appropriate options.

Explicit Commands

Generate podcast:

/gencast:podcast lecture.md

With custom style and audience:

/gencast:podcast api_docs.md --style interview --audience technical

Generate planning document only:

/gencast:plan research_paper.md

Styles

  • educational (default) - Structured learning format
  • interview - Q&A conversational style
  • casual - Relaxed, friendly discussion
  • debate - Contrasting perspectives

Audiences

  • general (default) - Broad accessibility
  • technical - Assumes domain knowledge
  • academic - Scholarly approach
  • beginner - Introductory level

Voices

6 voices available (alloy, echo, fable, onyx, nova, shimmer):

Defaults:

  • HOST1: nova (bright, energetic)
  • HOST2: echo (warm, friendly)

Custom voices:

/gencast:podcast doc.md --host1-voice onyx --host2-voice alloy

Examples

Educational Podcast (Default)

/gencast:podcast lecture_notes.md

Technical Interview

/gencast:podcast api_guide.md --style interview --audience technical

Beginner-Friendly Casual

/gencast:podcast intro.md --style casual --audience beginner

Academic Debate

/gencast:podcast paper.md --style debate --audience academic --with-planning

Multi-Chapter Course

/gencast:podcast ch1.md ch2.md ch3.md -o course.mp3

Options

Option Description Values
--style Podcast style educational, interview, casual, debate
--audience Target audience general, technical, academic, beginner
--with-planning Generate comprehensive plan first flag
--save-dialogue Save dialogue script to text file flag
--save-plan Save planning document to text file flag
-o, --output Output file path path (default: podcast.mp3)
--host1-voice Voice for HOST1 alloy, echo, fable, onyx, nova, shimmer
--host2-voice Voice for HOST2 alloy, echo, fable, onyx, nova, shimmer

Commands

/gencast:podcast

Generate podcast with full control over options.

/gencast:podcast <input-files...> [options]

/gencast:plan

Generate planning document only (no audio).

/gencast:plan <input-file>

Input Formats

Supported formats:

  • Markdown (.md)
  • Text (.txt)
  • PDF (.pdf) - Requires MISTRAL_API_KEY environment variable

Output

With --minimal flag, gencast shows:

  • Milestone: Planning (if --with-planning)
  • Milestone: Generating dialogue
  • Milestone: Synthesizing audio
  • Final: Output path and duration

Advanced Usage

With Planning and Dialogue Saving

/gencast:podcast complex_doc.md --with-planning --save-dialogue --save-plan

Creates:

  • podcast.mp3 - Audio file
  • podcast_dialogue.txt - Dialogue script
  • podcast_plan.txt - Planning document

Review Plan Before Audio

/gencast:plan document.md
# Review document_plan.txt
/gencast:podcast document.md --with-planning

Custom Voice Pairing

/gencast:podcast story.md --host1-voice fable --host2-voice shimmer --style casual

Reference Documentation

  • skills/gencast/references/voices.md - Voice options and pairings
  • skills/gencast/references/styles.md - Style/audience combinations

Tips

  1. Always uses --minimal - Context-friendly output
  2. Default works well - educational + general suits most content
  3. Match style to content - interview for Q&A, debate for analysis
  4. Use planning for complex docs - Ensures comprehensive coverage
  5. Save dialogue for review - Check script before re-generating

Requirements

  • gencast CLI (pip install gencast)
  • For PDFs: MISTRAL_API_KEY environment variable

License

MIT

Related skills

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.