Start

Overview

This skill enables text-to-speech for the current Claude session by creating a session-specific marker file. It uses TMPDIR (or /tmp) and CLAUDE_SESSION_ID to toggle audio delivery, so substantial responses can be read aloud as audio summaries.

How This Skill Works

It creates a marker file named tts-enabled-${CLAUDE_SESSION_ID} in ${TMPDIR:-/tmp}. The presence of this file signals the system to enable TTS for this session, and you can confirm by requesting an audio summary in future responses.

When to Use It

• You want audio summaries for long or complex responses.
• You’re working hands-free (e.g., coding, notes, or on a presentation).
• Accessibility needs require TTS for content consumption.
• You want quick spoken feedback on results or explanations.
• You want to review content without staring at the screen.

Quick Start

1. Step 1: Ensure TMPDIR is writable (e.g., export TMPDIR=/tmp).
2. Step 2: Enable TTS for this session: touch "${TMPDIR:-/tmp}/tts-enabled-${CLAUDE_SESSION_ID}"
3. Step 3: Request an audio summary or ask a question and listen for the spoken response.

Best Practices

• Verify TMPDIR is set and accessible, and that CLAUDE_SESSION_ID is available for the session.
• Run the exact touch command to create the marker file: touch "${TMPDIR:-/tmp}/tts-enabled-${CLAUDE_SESSION_ID}".
• Test by requesting an audio summary to confirm TTS is active.
• Keep the marker file only for the active session; delete it when you no longer need audio output.
• Be mindful of content length; for long docs, consider chunking prompts to maintain clear audio.

Example Use Cases

• Reading lengthy logs or reports aloud while debugging.
• Developing code with audio explanations of functions or algorithms.
• Delivering a spoken recap of meeting notes or design docs.
• Using TTS to assist users with accessibility needs during knowledge work.
• Reviewing research papers or articles on public transport without staring at a screen.

Frequently Asked Questions