
Once hands-free consent is acknowledged, EchoAssist keeps capture running and waits for one of your wake phrases. A matched phrase issues a live answer request, drafted in the first person so you can read it aloud as it is.
Listening cannot start until the hands-free consent version is acknowledged in section 01.
No draft yet. Start listening, then say or trigger a wake phrase to prepare a first-person answer.
Request metadata for 2026-10-05 is recorded in the shared voice request log; the daily counter resets at midnight UTC.
Detection runs on client-captured audio while listening is active. These phrases are shown here for reference; they are configured on the Wake Phrases page and are read-only on this page.
Between 1 and 5 phrases may be configured. Each phrase must be 2–6 words, letters and numbers only, up to 60 characters, and case-insensitively unique.
Edit wake phrasesHands-free runs on the same account and the same voice envelope as the rest of the instrument. These are the limits it holds to while it is listening, so you can see exactly what it will and will not do before you start.
ResetThe daily counter resets at midnight UTC. Requests used today can be read on the listening console above, which holds the live count.

A private, account-scoped voice and text assistant for one owner. Ask a question in text or request a live answer mid-conversation, grounded in your own uploaded background and custom instructions — never in invented experience.
Sign-in is delegated to the identity provider; your account record is created automatically on first sign-in.
These are the working limits of the assistant for one account owner. They are stated before sign-in and applied on every request.
6stated limits
02Lifecycle
Your message leaves the composer with a unique request ID.
One active request per account, held for 90 seconds.
The AI gateway is called once — 60-second timeout, zero retries.
Output is bounded and checked against the spoken-answer budget.
The answer is returned for you to read aloud.
03 Grounding
The assistant is allowed to be useful precisely because it is bounded. These are the sources it may draw on, and the things it will never do on your behalf.
What grounds an answer
Two sources, supplied as data rather than as instructions.
What the assistant will never do
Hard limits, stated plainly rather than discovered later.
With no profile attached, the assistant still helps with general knowledge and uses conditional language instead of making claims about your career.
EchoAssist — Echo
EchoAssist is private to a single signed-in account owner, and your conversation content is never persisted — the assistant drafts and returns, and nothing of the conversation is stored.
No comments yet. Be the first!