Agent skill
read-aloud
Generate a standalone HTML reader with Kokoro TTS audio and word-synced highlighting from any markdown file. Use when user wants to listen to an essay, proofread by ear, or create an audio reader.
Install this agent skill to your Project
npx add-skill https://github.com/leegonzales/AISkills/tree/main/ReadAloud/read-aloud
SKILL.md
Read Aloud
Generate a standalone HTML reader with high-quality TTS audio and real-time word highlighting from any markdown file.
When to Use
Invoke when user:
- Asks to "read this aloud" or "generate audio" for a markdown file
- Wants to proofread by ear (catch awkward phrasing, rhythm issues)
- Needs a shareable audio reader (single HTML file, no server needed)
- Uses
/read-aloudcommand
Prerequisites
- macOS with Apple Silicon (M1+)
- Python 3.13+ (required for MLX)
- ffmpeg (for MP3 compression)
Core Command
SKILL_DIR="path/to/read-aloud"
$SKILL_DIR/scripts/read-aloud.sh <markdown-file> [options]
First run auto-installs dependencies to ~/.read-aloud/venv/.
Options
| Flag | Default | Description |
|---|---|---|
--voice |
af_heart |
Kokoro voice variant |
--speed |
1.0 |
Speech rate (0.5-2.0) |
--output-dir |
~/.read-aloud/output/<slug>/ |
Custom output directory |
--strip-sections |
(none) | Comma-separated heading names to skip |
--no-open |
(off) | Don't open browser after generation |
Pipeline
Three-stage process (~2-5 min for a typical essay):
- Generate Audio — Kokoro TTS produces per-paragraph WAV chunks
- Align Words — Whisper transcribes each chunk for drift-free word timestamps
- Build Reader — Assembles standalone HTML with embedded MP3 + word sync
Output
Single reader.html file with:
- Base64-embedded MP3 audio (fully standalone, no server needed)
- Per-word highlighting synced to audio playback
- Sticky playback controls (play/pause, stop, speed, progress bar)
- Keyboard shortcuts (Space, Escape, Arrow keys)
- Click any word to seek
Available Voices
| Voice | Description |
|---|---|
af_heart |
American female, warm (default) |
af_bella |
American female, clear |
am_adam |
American male, deep |
am_michael |
American male, neutral |
bf_emma |
British female, elegant |
bm_george |
British male, distinguished |
Examples
# Basic usage — generates reader at ~/.read-aloud/output/my-essay/
$SKILL_DIR/scripts/read-aloud.sh ~/Documents/my-essay.md
# Custom voice and speed
$SKILL_DIR/scripts/read-aloud.sh post.md --voice bm_george --speed 1.1
# Output to specific directory
$SKILL_DIR/scripts/read-aloud.sh draft.md --output-dir ./preview/
# Strip specific sections before generating
$SKILL_DIR/scripts/read-aloud.sh post.md --strip-sections "Brief,Links & Resources"
References
- Load
references/voice-guide.mdfor detailed voice descriptions and speed tuning - Load
references/examples.mdfor worked examples with expected output - Load
references/troubleshooting.mdwhen errors occur
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
context-continuity
High-fidelity context transfer protocol for moving conversations between AI agents. Preserves decision tempo, open loops, and critical context with graceful degradation. Use when the user says "transfer," "handoff," "continue this in another chat," or needs to work around context window limits. Produces structured artifacts (Minimal ~200 words, Full ~1000 words). DO NOT trigger on simple "summarize our conversation" requests—only when transfer intent is explicit.
codex-peer-review
silicon-doppelganger
Build psychometrically accurate personal proxy agents for the PAIRL Conductor system. Extracts personality, decision heuristics, and values into portable schemas that enable AI agents to negotiate, filter, and act on a principal's behalf.
requesting-code-review
Use when completing tasks, implementing major features, or before merging to verify work meets requirements. Dispatches three independent reviewers in parallel.
fabric-patterns
Run danielmiessler/fabric CLI patterns for content analysis, extraction, summarization, writing, security analysis, and more. Use when user asks to "use fabric," "run a pattern," "extract wisdom," "summarize with fabric," or when piping content through AI patterns would be more effective than inline processing. Triggers include "fabric," "pattern," "extract wisdom," "summarize this article," "analyze this threat report," or any reference to a specific fabric pattern name.
moltbook-enclave
Secure, air-gapped interface for Moltbook (social network for AI agents). Isolates untrusted external content from your main agent's memory and context.
Didn't find tool you were looking for?