Sag

Name: Sag
Author: steipete

steipete/clawdis

3k installs
385k repo stars
Updated August 3, 2026
steipete/clawdis

sag is an agent skill for the sag CLI that turns text into ElevenLabs speech with voice, model, and delivery tag control.

About

sag is an agent skill for the sag CLI that provides mac-style say UX on top of ElevenLabs text-to-speech. It requires ELEVENLABS_API_KEY or SAG_API_KEY and supports quick commands such as sag "Hello there", sag speak with a named voice, sag voices, and sag prompting for model tips. Default model is eleven_v3 for expressive output, with eleven_multilingual_v2 for stability and eleven_flash_v2_5 for speed. Pronunciation guidance covers respelling, --normalize auto for numbers and URLs, --lang bias, v3 pause tags like [short pause], and v2 SSML break elements where supported. Voice replies for chat agents can write MP3 files with sag -v Clawd -o /tmp/voice-reply.mp3 and return MEDIA paths. Agents should confirm voice and speaker before long output and pick delivery tags such as [whispers] or [excited] for character. Install via brew steipete/tap/sag. Use when users request spoken responses, voice character delivery, or local TTS playback in OpenClaw style agent workflows.

ElevenLabs TTS via sag CLI with ELEVENLABS_API_KEY or SAG_API_KEY.
Models: eleven_v3 default, eleven_multilingual_v2 stable, eleven_flash_v2_5 fast.
v3 supports audio tags like [whispers] and [short pause]; v2 supports SSML break tags.
Voice reply pattern writes MP3 and returns MEDIA:/path for chat delivery.
Install with brew steipete/tap/sag; confirm voice before long output.

Sag by the numbers

3,031 all-time installs (skills.sh)
+160 installs in the week ending Aug 2, 2026 (Skillselion tracking)
Ranked #72 of 550 CLI & Terminal skills by installs in the Skillselion catalog
Security screen: LOW risk (skills.sh audit)
Data as of Aug 3, 2026 (Skillselion catalog sync)

At a glance

sag capabilities & compatibility

Capabilities: elevenlabs tts via sag speak and quick commands · voice listing and model selection across v3 and · pronunciation normalization and language bias fl · v3 audio delivery tags and v2 ssml break support · mp3 output for agent media voice replies
Works with: openai
Use cases: orchestration · transcription
Pricing: Bring your own API key
Requires keys: ELEVENLABS_API_KEY

From the docs

What sag says it does

MEDIA:/tmp/voice-reply.mp3

SKILL.md

npx skills add https://github.com/steipete/clawdis --skill sag

Add your badge

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

[![Listed on Skillselion](https://skillselion.com/badge/skills/steipete/clawdis/sag.svg)](https://skillselion.com/skills/steipete/clawdis/sag)

Installs	3k
repo stars	★ 385k
Security audit	3 / 3 scanners passed
Last updated	August 3, 2026
Repository	steipete/clawdis ↗

How do I make an agent speak responses locally with ElevenLabs voices and correct model-specific delivery rules?

Generate ElevenLabs speech with the sag CLI using voice selection, model-specific delivery tags, and local playback for agent voice replies.

Who is it for?

OpenClaw or agent setups needing quick ElevenLabs TTS from the sag CLI with voice character control.

Skip if: Skip when you need non-ElevenLabs TTS or cloud-hosted streaming without the sag binary installed.

When should I use this skill?

User asks for voice reply, spoken output, sag command, ElevenLabs TTS, or audio tag delivery like whispers or excited.

What you get

Played or saved MP3 audio with chosen voice, normalization, and model-appropriate pause or SSML tags.

Local MP3 speech output or played audio

By the numbers

Supports ELEVENLABS_API_KEY with SAG_API_KEY fallback
Installable via steipete/tap/sag Homebrew formula

Files

SKILL.mdMarkdownGitHub ↗

sag

Use sag for ElevenLabs TTS with local playback.

API key (required)

ELEVENLABS_API_KEY (preferred)
SAG_API_KEY also supported by the CLI

Quick start

sag "Hello there"
sag speak -v "Roger" "Hello"
sag voices
sag prompting (model-specific tips)

Model notes

Default: eleven_v3 (expressive)
Stable: eleven_multilingual_v2
Fast: eleven_flash_v2_5

Pronunciation + delivery rules

First fix: respell (e.g. "key-note"), add hyphens, adjust casing.
Numbers/units/URLs: --normalize auto (or off if it harms names).
Language bias: --lang en|de|fr|... to guide normalization.
v3: SSML <break> not supported; use [pause], [short pause], [long pause].
v2/v2.5: SSML <break time="1.5s" /> supported; <phoneme> not exposed in sag.

v3 audio tags (put at the entrance of a line)

[whispers], [shouts], [sings]
[laughs], [starts laughing], [sighs], [exhales]
[sarcastic], [curious], [excited], [crying], [mischievously]
Example: sag "[whispers] keep this quiet. [short pause] ok?"

Voice defaults

ELEVENLABS_VOICE_ID or SAG_VOICE_ID

Confirm voice + speaker before long output.

Chat voice responses

When the user asks for a "voice" reply (e.g., "crazy scientist voice", "explain in voice"), generate audio and send it:

# Generate audio file
sag -v Clawd -o /tmp/voice-reply.mp3 "Your message here"

# Then include in reply:
# MEDIA:/tmp/voice-reply.mp3

Voice character tips:

Crazy scientist: Use [excited] tags, dramatic pauses [short pause], vary intensity
Calm: Use [whispers] or slower pacing
Dramatic: Use [sings] or [shouts] sparingly

Default voice for Clawd: lj2rcrvANS3gaWWnczSX (or just -v Clawd)