Streamloop

Revokes an ingest key for good: the encoder using it is cut off at once. It cannot be restored; a new key for the same source: create_scene_ingest_key. Answers revoked (the key's id).

Read the scene's draft (or a published version, with version): a resource (with its revision and children, the sub-paths to read next), a type such as element/Text or source/data (its props), a mask ending in /* (a list), or "" (the types). as: "image" on a frame or layer → a picture of it. (In a scene's draft; scene: its id.) Answers summary and value (the resource, with its revision and children), or image for as "image"; plus draftRevision (version, for a published version).

Fetch a URL once, changing nothing in the show: HTTP status, content type, an outline of the JSON (keys, their values, array lengths) or the page's title and text, and the rows a table would get from it with connector, pick and headers — or, with connector mcp, an MCP server's tools, or one tool's rows with tool and args. Use it before creating a URL or MCP table, and to find the pick path or selector. (In a scene's draft; scene: its id.)

Delete whole resources (frames, layers, sources, components). All or nothing: anything in use by something not removed in the same call is refused, with what uses it. (In a scene's draft; scene: its id.) Removing a resource needs its revision in `revisions` (path → the revision you read; "*": whatever is there, on purpose); without it the answer is REVISION_REQUIRED. It changes the draft only: viewers see nothing until publish_scene, and discard_scene_draft brings back what was published. Answers summary, removed (the paths) and draftRevision.

Create, replace or change one resource; get <type> shows what each takes, with an example. value is the whole resource, as get answers it: a frame { code: "<JSX>" } or, as JSON, { name, root }; a component { description, props, tsx }; a source or table its fields. edits change part of it: [{ old, new }] replaces text in its code (copy old from what get answers: code is stored formatted), or an object is merged into its fields. One layer changes alone at its own path: frame/<id>/<layer> with edits { props: { … } }. Every set is checked as the studio checks it; a refused set changes nothing: read the error, fix, send again. The answer has the path it is stored at, its new revision, what changed, and draftRevision (the draft as it is now: what publish takes). (In a scene's draft; scene: its id.) Changing a resource that exists needs its `revision`, from the answer that read it ("*" overwrites whatever is there, on purpose); without it the answer is REVISION_REQUIRED. A new path needs none.

Set the window a stream runs in, once or on a repeat, and arm it. The stream must already be startable — a destination plus either a playable video playlist or an attached published scene (check_stream_readiness) — otherwise the window arrives and the start fails. repeat=daily re-arms the window exactly 24h later; repeat=weekly re-arms +7 days, or on the weekdays given in `days`. Both repeats REQUIRE scheduleEndAt and a window shorter than 24h; a one-off (repeat=none, the default) may be open-ended. `days` are UTC weekdays of the start instant, so a 9pm Monday Eastern window repeats on TUESDAY — the result echoes the committed window in UTC with its UTC weekday so this is visible. Replaces any existing window; cancel_stream_schedule disarms it. Returns the stream (id, name, state, scheduleStartAt, scheduleEndAt, scheduleRepeat, scheduleRepeatDays) and committedWindow: the window in UTC, its UTC start weekday, and the times exactly as you sent them.

Switch which workspace this connection works in, for every later tool call on this connection (other connections, even ones sharing the same login, are unaffected). Takes an `id` from list_workspaces; membership is checked here and again by the API on every request, so this grants no access by itself. Use it when a stream, upload or destination the user describes is not in this workspace. Overrides a workspace the client pinned in its own configuration (the X-Workspace-Id header, or MCP_WORKSPACE_ID on stdio) for the rest of this connection. Call it again at any time to switch back.

Saves an API key or token a scene's tables send (sealed; never shown again). A table names it as "$secret:<name>" in a header value or a URL's query. A name that exists already has its value replaced, and tables using it send the new one. Only a key the user gave you. Answers name and updatedAt.

Stop or resume sending to one of a stream's destinations without removing it. Disabling works while the stream is live — that output goes off the air at once — except for its last enabled destination (LAST_DESTINATION: stop_stream instead). Enabling is refused while live (STREAM_LIVE). Returns the stream's whole destination list after the change, as list_stream_destinations does.

Makes a stream play a scene (its published version) instead of a video playlist, or with scene null stops doing so. Refused (STREAM_LIVE) while the stream runs: stop it, set the scene, start it again. The scene must be published first (publish_scene).

Start streaming now, looping the stream's playlist (or playing its scene) until it is stopped. THIS PUTS VIDEO IN FRONT OF THE PUBLIC, so confirm with the user first unless they just asked for it. Requires a destination plus either a playable video playlist (at least one item) or an attached published scene (set_stream_scene) — check_stream_readiness reports what is missing without changing anything, and is worth calling first. A failed pre-flight is NOT an error: the call succeeds and the answer's `state` is `invalid` — read `state` first: invalid means refused, with `issues` saying why (fix them and start again); `preparing` means accepted. Media facts (duration, sound) arrive about 15-25 s after an upload completes, so a playlist of just-finished uploads can be refused until then. Other refusals (a state that cannot start, a guard failing at the moment of starting) are errors. Credits are NOT part of the pre-flight: the balance is verified when the stream actually activates, so an otherwise-ready stream can still fail to start on an empty wallet — get_billing_overview is what tells you that. A playlist loop takes the playlist as it currently is, with nothing to publish first; a scene plays its published version (publish_scene). Asynchronous: the stream reports a starting state immediately and is live within about a minute. To start at a future time (or every day) use schedule_stream instead. Returns state, started (false when refused), issues (when refused) and startStream (the stream: id, name, state).

Begin connecting a Google account so its YouTube channels can be streamed to with managed broadcasts (titles, privacy, thumbnails, live stats). Returns a Google consent URL — give it to the user to open in a browser; consent cannot be completed by an agent. Afterwards, list_youtube_channels shows the authorized channels and create_youtube_destination turns one into a destination. Also the fix for a destination whose status is needs_reauth or revoked. Not needed for plain RTMP publishing to YouTube — that is create_rtmp_destination, but it cannot manage the broadcast.

Stop a stream that is live or starting. It stops publishing and becomes inactive; nothing is deleted and it can be started again with the same configuration. Does NOT cancel a recurring schedule — a daily or weekly stream stopped here starts again at its next window, so call cancel_stream_schedule as well to keep it off.

Use this ONLY to put an edited playlist on the air while its stream is LIVE — the hot swap the product calls publishing. The stream changes over to the new arrangement without going off the air. NOT needed to set a stream up and NOT needed before a first start: starting a stream takes the playlist as it is then, so on a stream that is not live this tool just fails with NOT_LIVE. Asynchronous — newly added media may have to be encoded first, so poll get_playlist until the swap is no longer in progress. expectedDraftVersion is required (the `version` from get_playlist) so a concurrent edit is reported instead of going live by accident. Other failures name their cause: STALE_DRAFT (refetch and retry), EMPTY_PLAYLIST, BROKEN_ITEM (an item's media was deleted or never finished uploading), RECIPE_MISMATCH, NOT_FOUND.

Stop a playlist using a custom order; it returns to playing in its own item order (or shuffled, per the playlist's mode). The order itself is kept and can be assigned again. Like assigning, it takes effect on a running stream immediately, with no swap_live_playlist needed.

Maintain an existing destination: rename it, replace the stream key after the platform rotated it (custom RTMP only — a connected YouTube channel has no key to set), and archive or restore it. Archiving hides a destination from the picker without deleting it; it is reversible with archived=false. A live stream keeps publishing with the key it started with, so restart the stream after rotating a key. Pass only the fields to change. NOT transactional: each change commits on its own, and on a failure the response lists which ones already applied.

Change a stream's settings (name, quality, framerate, duration, storage, backup ingest) and/or its YouTube broadcast settings via the `youtube` object. Where it publishes is add_stream_destination / set_stream_destination_enabled / remove_stream_destination; when it runs is schedule_stream. The `youtube` settings only apply when one of the stream's destinations is a YouTube channel connected through Streamloop — not a custom RTMP destination, and not YouTube reached through a raw RTMP URL and key — and are otherwise skipped with a notice. Set retryMirror=true to re-attempt mirroring the schedule onto YouTube after a reconnect or a quota recovery. PARTIAL SUCCESS: settings are applied first, so if the YouTube settings then fail, `stream` still reflects the applied settings and `youtubeSettingsError` explains the YouTube failure — the settings change is NOT rolled back.