Which agents work with it?

Any agent that reads skills from a directory. Claude Code, Codex and Gemini CLI read it natively; Cursor and VS Code connect over MCP.

The voice runs on your own machine, so there is nothing to meter. The first run downloads it, about 340 MB.

Yes. The skill is MIT-licensed and the authoring format is public.

Wait for an answer · Bisque docs

GET /v1/presentations/{presentationId}/surveys/{surveyId}/response · API v1 · versioning

Returns the caller's own answer to one survey on one of their presentations. With wait, the request is held on a listener for up to that many seconds (at most 25) and answers the moment the survey is complete, or with until=submit the moment the viewer presses Continue (answers save as they are typed, so a complete text answer can still be half-written). On timeout it returns the answer as it stands with 200, and the caller polls again. This is the wait an agent already does for a deploy or a CI run, shaped like gh run watch. API-key callers only.

Authentication

Send an API key as Authorization: Bearer bisque_live_…. Keys come from bisque.cloud/setup/keys; see Authentication.

Request

Parameter	In	Meaning
`presentationId`	path, required
`surveyId`	path, required	The `id` in the slide's `data-survey-spec`.
`wait`	query	Seconds to hold the request before answering with the answer as it stands. Default 0.
`until`	query	`complete` (default) answers once every required question has an answer. `submit` waits for the viewer to press Continue, or play on the survey's slide.

Example

curl -X GET https://bisque.cloud/v1/presentations/{presentationId}/surveys/{surveyId}/response \
  -H "Authorization: Bearer $BISQUE_API_KEY"

Response

200: { presentationId, surveyId, complete, submitted, answers?, submittedAt?, updatedAt? }; answers maps question id to value when complete.

Errors:

400: wait is not a number of seconds, or until is not complete or submit.
401: Missing or invalid credential.
404: Not one of your presentations, or no such survey.

Other endpoints

Page	Call	What it does
Create a presentation	`POST /v1/presentations`	Put the slides up now and narrate them one at a time.
Get a presentation	`GET /v1/presentations/{presentationId}`	The open creation, to resume it from a new session.
Add slides	`PATCH /v1/presentations/{presentationId}`	Append sections while the creation is open.
Update a slide	`PATCH /v1/presentations/{presentationId}/slides/{slideKey}`	Give one slide its narration after the MP3 is uploaded.
Finalize	`POST /v1/presentations/{presentationId}:finalize`	Claim the version. Slides still without audio stay silent.
Read as text	`GET /api/presentations/context`	Any shared presentation as a transcript. No credential needed.