- Home
- Skills
- Technical Documentation
- DocuVoice: README to Podcast
More screenshots
Works with the AI Agents you already use
DocuVoice: README to Podcast
Turns a README or Markdown document into a local multi-voice MP3. Classifies sections, narrates them with ElevenLabs, and writes an MP3 plus a timeline. Does not host or upload the audio.
$1
DocuVoice: README to Podcast
Example session with this skill installed
Convert this GitHub README to a podcast: https://github.com/pablomg-dev/docu-voice. Use multi-voice mode with sound transitions.
- Read your context and instructions
- Compiled the docuvoice
I have processed the README from pablomg-dev/docu-voice. The content was split into 5 sections and rewritten for natural speech. Total character count is 1,420, with an estimated cost of 1,620 ElevenLabs credits.
Narration Result
The audio has been generated and saved locally.
| Property | Value |
|---|---|
| MP3 Path | /Users/dev/projects/docu-voice/podcast.mp3 |
| Timeline Path | /Users/dev/projects/docu-voice/podcast.timeline.json |
| Mode | Multi-voice |
| Transitions | Enabled (whoosh chime) |
| Assembly | ffmpeg (Re-encoded mono) |
| Language | English |
Audio Timeline
| Start Time | Section Title | Type | Voice |
|---|---|---|---|
| 00:00 | Introduction | introduction | Jessica |
| 00:45 | What you need | steps | Bella |
| 01:20 | Get the document | code | Adam |
| 02:05 | Permissions & data | warning | Antoni |
| 02:50 | Where this skill stops | introduction | Jessica |
Next steps
- Open the MP3 file in your preferred player to verify the narration flow.
- Check the
.timeline.jsonfile if you need to build a custom player UI. - Install
ffmpegif you notice the audio segments are byte-joined without transitions.
Connects securely to your tools. The creator never sees your data.
About this skill
Reading long documentation is time-consuming and often ignored. This skill converts GitHub READMEs, Markdown files, or pasted technical text into narrated MP3 podcasts for offline listening.
What it does
- Section Classification scans Markdown headers to categorize content as introduction, code, warning, or steps.
- Natural Narration converts technical syntax like code blocks and lists into human-readable spoken sentences.
- Multi-Voice Casting assigns distinct ElevenLabs voices to different section types (e.g., Adam for code, Bella for steps).
- Audio Assembly generates a local MP3 file with optional sound-effect transitions between sections.
- Timeline Generation creates a JSON index of section start times for easy navigation.
How it works
- Provide Document by passing a GitHub URL, a local Markdown file path, or pasting text directly.
- Review Estimate to see the character count and projected ElevenLabs credit usage before processing.
- Execute Narration using the Node.js script to call the ElevenLabs API and assemble the audio locally.
- Access Output to receive the final MP3 path and the structured timeline data.
Frameworks & tools
Built for Node.js 18+ environments. It utilizes the ElevenLabs API (eleven_turbo_v2_5) for high-quality speech synthesis and optionally uses ffmpeg for advanced audio joining and sound transitions.
Why this beats prompting it yourself
Manual prompting often results in literal readings of code symbols like "backtick backtick backtick" or skipped warnings. This skill implements a specific classification heuristic and a speakable-text rewriter that ensures technical documentation sounds natural to the human ear.
Use cases
- Convert library documentation into audio for consumption during commutes.
- Generate multi-voice audio briefings from software release notes.
- Audit the clarity of a tutorial by listening to its narrated flow.
- Create accessible audio versions of GitHub READMEs for offline learning.
Known limitations
Does not host or upload audio files. Requires a personal ElevenLabs API key. Narration is limited to 50 sections and 40,000 characters per run.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Technical Documentation.
Requires Node.js 18+ and ElevenLabs API key. ffmpeg required for transitions. No hosting or cloud upload. Single-voice narrator recommended for non-English text.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
Trust & safety
Security scan
Security scan passed 3 days ago
- Software
Node.js