myvoice

Draft outbound in the user's own voice AND keep that voice canon current. DRAFT mode (default): use before writing, rewriting, polishing, or replying to emails, DMs, LinkedIn messages, follow-ups, intros, product updates, scheduling notes, warm asks, paid-pilot framing, cold messages, OR any spoken script the user will say out loud (Looms, VSLs, video VO, call openers) where they want it to sound like a real person, not an assistant. The voice composes on two axes: a MODE (warm founder vs confident sales) and a CHANNEL (typed vs spoken). Triggers: draft in my voice, write this email/DM/cold message, script this Loom/VSL, and "in my confident/sales voice" or "in my warm voice" to force a mode. LEARN/CAPTURE mode: when the intent is learn my voice, capture my voice, remember how I talk, update my voice, or harmonize my voice, fold an approved draft or correction into the voice canon. Always enforce the user's punctuation preference: never use em dashes.