AgentSkillsCN

sag

采用 Mac 风格的“say”用户体验，实现 ElevenLabs 文本转语音功能。

SKILL.md
--- frontmatter
name: sag
description: ElevenLabs text-to-speech with mac-style say UX.
homepage: https://sag.sh
metadata: {"crocbot":{"emoji":"🗣️","requires":{"bins":["sag"],"env":["ELEVENLABS_API_KEY"]},"primaryEnv":"ELEVENLABS_API_KEY","install":[{"id":"brew","kind":"brew","formula":"steipete/tap/sag","bins":["sag"],"label":"Install sag (brew)"}]}}

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

Audio Tags are inline directives in square brackets [] that add emotional nuance, tone shifts, non-verbal sounds, and effects to generated speech. Place them anywhere in the text—before, after, or mid-sentence.

Syntax:

  • •Use lowercase for consistency (case-insensitive)
  • •Combine tags like [nervous][whispers] for layered effects
  • •Tags persist until overridden by a new tag or the segment ends
  • •No explicit terminator like [/tag] exists
  • •Punctuation like ellipses (…) or CAPS boosts emphasis alongside tags
  • •Best results with prompts >250 characters

Tag Categories:

CategoryExamples
Emotions[happy], [sad], [angry], [curious], [sarcastic], [excited], [nervous]
Delivery[whispers], [shouts], [sings], [strong French accent]
Non-verbal[laughs], [sighs], [clears throat], [breathing heavily], [exhales]
Pacing[pause], [short pause], [long pause]
Effects[gunshot], [applause], [explosion] (experimental)

Examples:

bash
sag "[whispers] This is secret... [excited] But it works!"
sag "[nervous][whispers] I'm not sure about this. [pause] Ok, let's do it."
sag "[sarcastic] Oh wow, that's SO impressive. [laughs]"

Notes:

  • •Works best with Creative or Natural stability settings
  • •Results vary by voice training data
  • •v3 does NOT support SSML <break> tags; use [pause] instead

Voice defaults

  • •ELEVENLABS_VOICE_ID or SAG_VOICE_ID

Confirm voice + speaker before long output.

Chat voice responses

NOTE: sag skill vs Auto-TTS This skill is for AGENT-INVOKED voice generation (when the AI explicitly runs sag). This is SEPARATE from the built-in Auto-TTS system (messages.tts.auto in config).

  • •sag skill: AI explicitly generates voice via sag CLI command
  • •Auto-TTS: Automatic voice conversion of all replies (see src/tts/tts.ts) Both should use the same voice ID for consistency.

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

bash
# Generate audio file (use voice ID directly for consistency with auto-TTS)
sag --voice-id ZD29qZCdYhhdqzBLRKNH -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: ZD29qZCdYhhdqzBLRKNH ("Female Humanoid - Futuristic") https://elevenlabs.io/app/voice-library?voiceId=ZD29qZCdYhhdqzBLRKNH