# Myvoice

> Consented local voice cloning with F5-TTS in the Promptus desktop app — microphone capture, reference preflight, fail-closed signal and word-accuracy gates, and a recorded human listening verdict before anything is called accepted. "myvoice" is the short callable name for the promptus-clone-voice skill; use it when the user wants to clone a voice, add narration, or generate speech locally. Do not use for cloud TTS or non-consented impersonation.

- Skill: `marroccofella/myvoice` (Agent Skill)
- Install (CLI): `npx skillmds@latest add marroccofella/myvoice`
- Raw SKILL.md: https://api.skillmd.com/api/skills/marroccofella/myvoice/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: DevOps & Infra
- Author: marroccofella (https://skillmd.com/u/marroccofella)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/marroccofella/myvoice

---


# myvoice → promptus-clone-voice

`myvoice` is the short, callable name for the **[promptus-clone-voice](../promptus-clone-voice/SKILL.md)** skill. Follow that skill's full protocol — it is the canonical source.

**Safety gate (non-negotiable):** confirm the user owns the voice or has the speaker's explicit permission; require `--consent-confirmed`; pass the fail-closed signal and word-accuracy gates; and record a human listening verdict before calling any clone accepted. Label shared synthetic audio as synthetic. Everything runs locally in Promptus — nothing is uploaded for generation.

See [promptus-clone-voice/SKILL.md](../promptus-clone-voice/SKILL.md) for the complete workflow, Windows paths, and reference-admission rules.