WebSocket API

SmoredBoard Documentation

Connect to the SmoredBoard WebSocket API to trigger sounds, listen for events, and integrate SmoredBoard with your own apps and tools.

Overview

The SmoredBoard WebSocket API lets external apps connect to a running SmoredBoard instance to trigger sounds, query state, and react to playback events in real time. Every connection is authenticated per-message with a token, and every response follows a consistent envelope shape.

Connection

Connect to the WebSocket server at:

ws://127.0.0.1:8181

The server binds to the loopback interface only, so the API is reachable from the same machine that's running SmoredBoard. localhost works as well, but 127.0.0.1 is the canonical address.

NOTE: The API is local-only by design, it will not respond to connections from other machines on the network.

NOTE: The user can disable the WebSocket API entirely under General Settings → WebSocket API. If the WebSocket fails to connect at all (rather than returning an authentication error), the API is most likely turned off, ask your user to re-enable it in their settings.

Authentication

Every message sent to the server must include a Token field. Authentication is checked per-message, there is no separate handshake or login step.

Token: "SMORED1999VERYGOODANDCOOL"

If the token is missing or wrong, the server replies with a stable error envelope you can detect programmatically:

{
  "Success": false,
  "Action": "Invalid Authentication",
  "Message": "Invalid authentication token.",
  "ErrorCode": "INVALID_TOKEN"
}

The Action field on auth failures is set to "Invalid Authentication" so existing handlers (like the Stream Deck plugin's authentication fail handler) can branch on it directly without parsing the message string.

Origin Gating

On top of the per-message token check, browser clients also pass through an Origin check on connect. Native clients (Stream Deck plugins, custom desktop apps, and anything that doesn't send an Origin header) are unaffected, this only applies to connections coming from a browser.

Client / Origin	Behavior
Native client (no `Origin` header)	Auto-allowed. Token check on each message still applies.
`Origin: "null"` (sandboxed iframes, `file://`, opaque origins)	Auto-allowed. Browsers send the literal string `"null"` for sandboxed contexts, gating on it would just block legitimate local-tool integrations without adding security.
`https://smoredboard.com` and any `*.smoredboard.com` over HTTPS	Auto-allowed (hardcoded).
Any other browser origin (first connection)	User is prompted with an Allow / Block dialog. The decision is persisted.
Origin previously approved	Allowed silently.
Origin previously blocked	Socket is closed immediately on connect.
`Origin` header present but not a parseable absolute URI	Blocked.

Origin gate status messages

When a browser connects with an origin that's not pre-approved, the server pushes a status message immediately after the WebSocket opens, so the client knows why nothing else is happening. Both messages use the standard error envelope:

// First connection from a new origin, dialog is up on the user's screen
{
  "Success": false,
  "Action": "PendingApproval",
  "ErrorCode": "PENDING_APPROVAL",
  "Message": "Waiting for the SmoredBoard user to approve this origin. The connection will stay open until they decide."
}

// Origin previously blocked, just blocked by the user, dialog dismissed,
// or the approval dialog couldn't be shown
{
  "Success": false,
  "Action": "OriginBlocked",
  "ErrorCode": "ORIGIN_BLOCKED",
  "Message": "This origin is blocked from connecting to SmoredBoard. The user can unblock it in General Settings → WebSocket API."
}

NOTE: The Message text on OriginBlocked varies depending on which path triggered it (pre-existing block, user just clicked Block, user dismissed the dialog with X, or the dialog couldn't be shown). Branch on ErrorCode: "ORIGIN_BLOCKED", not the message string.

Behavior after each message:

PendingApproval: connection stays open, but client messages are dropped while the dialog is up. If the user clicks Allow, normal request processing starts and the next message you send works as expected. If they click Block (or dismiss with X), the server follows up with an OriginBlocked message and closes the socket.
OriginBlocked: server closes the socket immediately after sending. Don't retry, reconnecting from the same origin will hit the same block.

NOTE: Don't assume onopen means “ready to use.” Watch for PendingApproval / OriginBlocked on the first message after connect, and only consider the connection usable once you've either gotten a successful response to a probe (e.g. Hello) or you're sure no gate message is coming.

Origin canonicalization

Origins are normalized to scheme://host[:nondefaultport], lowercased, before being compared against the approval list. So https://Foo.com and https://foo.com map to the same approval. Default ports are only stripped for http (:80) and https (:443); for any other scheme the port is preserved as-is.

Managing approvals

Users can review, revoke, and re-approve origins at any time under General Settings → WebSocket API in SmoredBoard.

Response Envelope

Every response from the server, uses the same outer shape:

{
  "Success": true,
  "Action": "<action name>",
  "Message": "<human-readable status or detail>",
  // ...action-specific fields
}

Success: true on success, false on any error.
Action: the action this response is for (matches the request's Action, or the event name for server-pushed messages).
Message: a human-readable string with detail about the result.
Deprecation (optional): a human-readable warning string that signals the request hit a legacy code path that may be removed in a future version. The field is omitted from the JSON entirely when there's nothing to warn about, so existing clients see no wire-format change.
ErrorCode (optional, error responses only): a stable string constant (e.g. "INVALID_TOKEN", "PROFILE_NOT_FOUND") intended for programmatic branching. Pair it with Message, which is the human-readable explanation. Omitted on success responses. See the Errors section for the full reference.
RequestId (optional): if your request includes a RequestId field, the server echoes the same value back on the response so multiple in-flight requests on the same connection can be matched up. Omitted entirely when you didn't send one.
Action-specific fields appear alongside these, e.g. GetProfiles returns a list of profiles, etc.

NOTE: Field names use PascalCase (Success, Message, Action, Token), this applies to both requests and responses.

NOTE: Deprecation currently only fires for profile-aware actions called with ProfileName instead of ProfileGuid (see the Requests section). Watch for it in your logs to catch usage of deprecated code paths early.

Requests

A request is a JSON object with at minimum an Action field naming the action to perform, and a Token field for authentication.

{
  "Action": "PlaySound",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "SoundPath": "C:\Users\You\Sounds\airhorn.mp3",
  "ProfileGuid": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c"
}

Correlating requests and responses

Requests can include an optional RequestId field of any string value. If present, the server echoes it back on the response so multiple in-flight requests on the same connection can be matched up by ID. The server doesn't care about the format, use whatever works for your client (UUID, monotonic counter, etc.).

// Request
{
  "Action": "GetProfiles",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "RequestId": "req-42"
}

// Response
{
  "Success": true,
  "Action": "GetProfiles",
  "Profiles": [ ... ],
  "RequestId": "req-42"
}

Identifying profiles and sounds

Sounds are addressed by their sound path, plus the profile they live in. There are two ways to identify a profile:

ProfileGuid: preferred. Stable across profile renames.
ProfileName: legacy fallback. Will break if the user renames the profile.

NOTE: Always prefer ProfileGuid in production integrations, ProfileName will silently stop matching the moment a user renames the profile. Any request sent using ProfileName instead of a ProfileGuid will include a Deprecation field (see Response Envelope), so you can detect this in logs even before a rename breaks you.

Events

Events are messages the server pushes to all connected clients without being requested. They use the standard response envelope, with Action set to the event name.

SoundFinished

Fired whenever a sound finishes playing in SmoredBoard.

{
  "Success": true,
  "Action": "SoundFinished",
  "SoundPath": "C:\path\to\sound.mp3"
}

SoundFinished is currently the only server-pushed event. New events will be added here as they ship.

Managing subscriptions

Every connected client is subscribed to every event by default. To opt out of a specific event, send the matching Unsubscribe<EventName> action. The opt-out is per-connection, closing and reopening the WebSocket resets the client to the default subscribed state.

NOTE: Because SoundFinished is the only event today, the only subscribe/unsubscribe pair available is the one below. Future events will ship with matching Subscribe<Name> / Unsubscribe<Name> actions.

UnsubscribeSoundFinished

Added in 1.0.4.5

Stop sending SoundFinished events to this connection. No-op if already unsubscribed.

Success response

{
  "Success": true,
  "Action": "UnsubscribeSoundFinished",
  "Message": "No longer receiving SoundFinished notifications."
}

SubscribeSoundFinished

Added in 1.0.4.5

Re-subscribe to SoundFinished events on this connection (the default state). No-op if already subscribed.

Success response

{
  "Success": true,
  "Action": "SubscribeSoundFinished",
  "Message": "Subscribed to SoundFinished notifications."
}

Errors

Errors share the same envelope as successful responses, there is no separate error message type. On failure, Success is false, Message describes what went wrong in human terms, and ErrorCode is a stable string constant you can branch on programmatically.

{
  "Success": false,
  "Action": "Invalid Authentication",
  "Message": "Invalid authentication token.",
  "ErrorCode": "INVALID_TOKEN"
}

Always check Success before trusting any action-specific fields in the response. For programmatic logic, branch on ErrorCode rather thanMessage, the message text can change between versions; the code is stable.

Error codes

The server emits a fixed set of stable ErrorCode values. The full list is below, grouped by category.

ErrorCode	When it fires
`INVALID_TOKEN`	The `Token` field is missing or doesn't match. Pairs with `Action: "Invalid Authentication"`.
`INVALID_REQUEST`	Catch-all for any unhandled exception during request dispatch (malformed JSON, unexpected server-side errors before action-specific handling takes over, etc.). Response `Action` is set to `"Error"`.
`UNKNOWN_ACTION`	The `Action` value isn't recognized by this server version. Often means the client is newer than the server, or the action was renamed.
`PENDING_APPROVAL`	Connection arrived from a browser origin that isn't pre-approved, the user is being shown an Allow / Block dialog. The connection stays open while you wait. Pairs with `Action: "PendingApproval"`. See the Origin Gating section.
`ORIGIN_BLOCKED`	The browser origin is blocked from connecting (previously rejected, or the dialog couldn't be shown). Server closes the socket immediately after sending. Pairs with `Action: "OriginBlocked"`.
`MISSING_PARAMETER`	A required field on the request was not provided. `Message` names which one (e.g. `"SoundPath is required"`, `"ProfileGuid or ProfileName is required"`).
`INVALID_PARAMETER`	A field was present but its value isn't acceptable (e.g. `"Fade time cannot be negative"`).
`PROFILE_NOT_FOUND`	The `ProfileGuid` / `ProfileName` on the request doesn't resolve to an existing profile.
`SOUND_NOT_FOUND`	The supplied `SoundName` / `SoundPath` doesn't resolve to a sound in the target profile.
`VOICE_CHANGER_NOT_FOUND`	The supplied `VoiceChangerId` doesn't match any built-in or custom voice changer (e.g. on `SetVoiceChangerById`).
`AUTOMATION_NOT_FOUND`	The supplied `AutomationId` on `TriggerAutomation` doesn't resolve to an existing automation.
`STREAM_NOT_FOUND`	The supplied `StreamId` on a Stream Chat command (e.g. `PauseStreamTriggers`) doesn't match any configured Twitch or YouTube chat stream. Use `GetAllChatStreams` to enumerate valid IDs.
`DEVICE_NOT_FOUND`	The audio device referenced on the request doesn't resolve to a known input/output device.
`NOT_AVAILABLE`	A required subsystem isn't available right now (e.g. `"AudioManagerV4 not available"`, `"Loopback manager not available"`). Often transient, especially during app startup, retry once before failing hard.
`PRO_REQUIRED`	The action is Pro-gated and the current user isn't a Pro subscriber. Surface an upsell to the user, don't silently swallow.
`RATE_LIMITED`	The connection exceeded the per-socket rate limit (50 requests/second). The offending message is dropped, see the Rate limiting section below for details.
`OPERATION_FAILED`	The operation ran but didn't produce a useful result for reasons that aren't a hard error (e.g. `"Jitter benchmark returned null (mic may not be on or already running)."`).
`INTERNAL_ERROR`	The server caught an unexpected exception while handling the request. `Message` contains the exception detail when safe to expose. Treat as transient and retry once before failing hard.

Rate limiting

Each connected socket is capped at 50 requests per second. The limit is per-socket (so two clients can each use their own 50/sec quota in parallel) and exists purely to prevent runaway spam, normal integrations won't come close to it.

When the cap is exceeded, the offending message is dropped (not queued, not delayed) and the server replies with:

{
  "Success": false,
  "Action": "RateLimited",
  "ErrorCode": "RATE_LIMITED",
  "Message": "Rate limit exceeded (50 req/sec). Message dropped."
}

NOTE: The rate-limit window resets when you reconnect.

Action Reference

Each of these actions wraps the canonical SmoredBoard Actions system, the same code path used by chat triggers, automations, and the in-app action chain editor. They all live behind the standard Token auth and the origin gate covered above.

Jump to a category: Sounds · Profiles · Microphone · Voice Changer · Censor · Sound Snap · Random Sound Mode · SFX Output · Hear Self · Keybinds · Automations · Chat Triggers · Diagnostics

NOTE: Pro-gated actions (marked with the Pro badge below) return a Success: false response with ErrorCode: "PRO_REQUIRED" and Message: "Pro required" for non-Pro users. Branch on ErrorCode === "PRO_REQUIRED" and surface an upsell prompt to the user.

Sounds

PlaySound

Added in 1.0.4.5

Play a sound from any profile. Pressing the same sound while it's playing stops it.

Request fields

SoundPath (string, required), full file path to the sound.
ProfileGuid (string, preferred) or ProfileName (string, fallback), which profile the sound belongs to. One of the two is required.

Example request

{
  "Action": "PlaySound",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "SoundPath": "C:\Users\You\Sounds\airhorn.mp3",
  "ProfileGuid": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c"
}

Success response

{
  "Success": true,
  "Action": "PlaySound",
  "Message": "Playing",
  "SoundPath": "C:\Users\You\Sounds\airhorn.mp3",
  "ProfileGuid": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c"
}

Errors

MISSING_PARAMETER, no SoundPath provided, or no ProfileGuid / ProfileName provided.
SOUND_NOT_FOUND, the file at SoundPath doesn't exist on disk.
INTERNAL_ERROR, unexpected exception during playback. Message contains the exception detail.

NOTE: If ProfileGuid / ProfileName doesn't resolve to an existing profile, the server still returns Success: true and silently does nothing. Validate profile references with GetProfiles before calling PlaySound if you need to detect bad references.

StopSound

Added in 1.0.4.5

Stop a specific sound by name in a specific profile.

Request fields

SoundName (string, required), name of the sound (custom name or filename without extension).
ProfileGuid (preferred) or ProfileName, required.

Example request

{
  "Action": "StopSound",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "SoundName": "airhorn",
  "ProfileGuid": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c"
}

Success response

{
  "Success": true,
  "Action": "StopSound",
  "Message": "Stopped",
  "SoundName": "airhorn",
  "ProfileGuid": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c",
  "ProfileName": "Gaming"
}

PlayRandomSoundPro

Added in 1.0.4.5

Play a random sound from a profile.

Request fields

ProfileGuid / ProfileName omitted, the current active profile is used.
A specific ProfileGuid (preferred) or ProfileName, plays a random sound from that specific profile.
Special: passing "__RANDOM__" as either ProfileGuid or ProfileName picks a random profile, then a random sound within it.
Special: passing "__ALL__" as either ProfileGuid or ProfileName plays one random sound from a flat pool of every sound across all profiles (differs from "__RANDOM__", which scopes to one randomly chosen profile).

Example request

{
  "Action": "PlayRandomSound",
  "Token": "SMORED1999VERYGOODANDCOOL"
}

Success response

{
  "Success": true,
  "Action": "PlayRandomSound",
  "Message": "Random sound triggered"
}

RandomSoundButtonPro

Added in 1.0.4.5

Play one random sound. With no profile fields this mirrors a user clicking the in-app Random Sound button (plays from the current active profile). Optionally accepts a profile reference to scope or randomize which profile the sound comes from.

Request fields

ProfileGuid / ProfileName omitted, the current active profile is used.
A specific ProfileGuid (preferred) or ProfileName, plays a random sound from that specific profile.
Special: passing "__RANDOM__" as either ProfileGuid or ProfileName picks a random profile, then a random sound within it.
Special: passing "__ALL__" as either ProfileGuid or ProfileName plays one random sound from a flat pool of every sound across all profiles (differs from "__RANDOM__", which scopes to one randomly chosen profile).

Success response

{
  "Success": true,
  "Action": "RandomSoundButton"
}

Errors

PRO_REQUIRED, the current user isn't a Pro subscriber. Message is "Pro required".
INTERNAL_ERROR, unexpected exception while picking or playing the sound. Message contains the exception detail.

NOTE: RandomSoundButton and PlayRandomSound are now functionally equivalent (both Pro, and both accept the same profile references and sentinels). RandomSoundButton with no fields specifically mirrors the in-app button, which is handy for UI-mirror integrations. Passing ProfileName instead of ProfileGuid may return a Deprecation field (see Response Envelope).

StopAllSounds

Added in 1.0.4.5

Stop all currently-playing sounds, respecting the user's fade-out setting.

Success response

{
  "Success": true,
  "Action": "StopAllSounds"
}

StopAllSoundsImmediately

Added in 1.0.4.5

Force-stop all sounds immediately, ignoring any fade-out setting.

Success response

{
  "Success": true,
  "Action": "StopAllSoundsImmediately",
  "Message": "All sounds stopped immediately"
}

FadeOutStopAllSounds

Added in 1.0.4.5

Stop all sounds with a custom fade-out duration. Does not modify the saved fade-out setting.

Request fields

FadeTime (number, required), fade duration in seconds. 0 stops immediately. Negative values are rejected.

Example request

{
  "Action": "FadeOutStopAllSounds",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "FadeTime": 2.5
}

Success response

{
  "Success": true,
  "Action": "FadeOutStopAllSounds",
  "Message": "All sounds fading out over 2.5s"
}

(Or "All sounds stopped immediately" if FadeTime is 0.)

Errors

INVALID_PARAMETER, FadeTime is negative.
NOT_AVAILABLE, the audio subsystem isn't ready (e.g. briefly during app startup). Retry once before failing hard.

RequestButtonDataPro

Added in 1.0.4.5

Fetch the current contents of a single soundboard slot (sound or voice changer) by its index. Used by auto-loading clients that mirror the in-app button grid. The image is returned as a base64-encoded string (compressed) so the response can be rendered directly in a browser.

Request fields

Index (number, required), zero-based slot index in the active profile.

Example request

{
  "Action": "RequestButtonData",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "Index": 3
}

Success response (sound slot)

{
  "Success": true,
  "Action": "AutoLoadButtonUpdate",
  "Message": "3 /-+-/ airhorn |-+-| C:\\Users\\You\\Sounds\\airhorn.mp3 |+| <base64-image-data>"
}

Success response (voice changer slot)

{
  "Success": true,
  "Action": "AutoLoadButtonUpdate",
  "Message": "3 /-+-/ Robot |-+-| __VOICECHANGER__ |+| <base64-image-data> |VC| <voice-changer-id>"
}

NOTE: The response Action is "AutoLoadButtonUpdate", not "RequestButtonData". The Message uses /-+-/, |-+-|, |+|, and |VC| as field delimiters. Voice-changer images are circular-masked; sound images are not.

Errors

PRO_REQUIRED, the current user isn't a Pro subscriber. Message is "{Index} /-+-/ Invalid Index or No Profile Loaded".
NOT_AVAILABLE, the main window isn't ready.
INDEX_OUT_OF_RANGE, the supplied Index doesn't resolve to a valid slot (no sound or voice changer at that position).

Profiles

GetProfiles

Added in 1.0.4.5

List all profiles available in the app.

Success response

{
  "Success": true,
  "Action": "GetProfiles",
  "Profiles": [
    { "name": "Gaming", "Id": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c", "ImagePath": "C:\\path\\to\\gaming.png" },
    { "name": "Streaming", "Id": "a1b2c3d4-...", "ImagePath": "" }
  ]
}

NOTE: Use the Id field as ProfileGuid in subsequent calls. It's stable across renames. ImagePath is the local file path of the profile's image (empty string if none); for raw image data, call GetProfileImage with the Id.

GetProfileSounds

Added in 1.0.4.5

List every sound file in a profile, with its name, full path, and button-slot index.

Request fields

ProfileGuid (preferred) or ProfileName, required.

Example request

{
  "Action": "GetProfileSounds",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "ProfileGuid": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c"
}

Success response

{
  "Success": true,
  "Action": "GetProfileSounds",
  "ProfileGuid": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c",
  "ProfileName": "Gaming",
  "Sounds": [
    { "Index": 0, "Name": "airhorn", "SoundPath": "C:\Users\You\Sounds\airhorn.mp3" },
    { "Index": 2, "Name": "applause", "SoundPath": "C:\Users\You\Sounds\applause.wav" }
  ]
}

Notes

Index is the button-slot index in the profile. Slots that hold voice changers, are empty, or point to a sound file that no longer exists on disk are skipped, that's why indices may have gaps.
Name is the user's custom name if set, otherwise the filename without extension.
SoundPath can be passed directly to PlaySound or StopSound.

Errors

PROFILE_NOT_FOUND, the supplied ProfileGuid / ProfileName doesn't resolve, or neither was provided.

GetProfileVoiceChangers

Added in 1.0.4.5

List every voice changer assigned to a button slot in a profile.

Request fields

ProfileGuid (preferred) or ProfileName, required.

Example request

{
  "Action": "GetProfileVoiceChangers",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "ProfileGuid": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c"
}

Success response

{
  "Success": true,
  "Action": "GetProfileVoiceChangers",
  "ProfileGuid": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c",
  "ProfileName": "Gaming",
  "VoiceChangers": [
    { "Index": 1, "VoiceChangerId": "builtin:robot1", "Name": "Robot", "IsBuiltIn": true },
    { "Index": 3, "VoiceChangerId": "1be13752690e4883a4ab92a642aff588", "Name": "MyCustomVC", "IsBuiltIn": false },
    { "Index": 5, "VoiceChangerId": "__NONE__", "Name": "None", "IsBuiltIn": false }
  ]
}

Notes

Index is the button-slot index. Sound slots are skipped, so indices have gaps and aren't sequential.
VoiceChangerId can be passed directly to SetVoiceChangerById.
IsBuiltIn: true means a SmoredBoard-provided voice changer; false means a user-created custom one.
The "__NONE__" entry is a “no voice changer” slot, selecting it via SetVoiceChangerById is equivalent to TurnOffVoiceChanger.

Errors

PROFILE_NOT_FOUND, the supplied ProfileGuid / ProfileName doesn't resolve, or neither was provided.

LoadProfile

Added in 1.0.4.5

Switch the active profile.

Request fields

ProfileGuid (preferred) or ProfileName, one is required.

Example request

{
  "Action": "LoadProfile",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "ProfileGuid": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c"
}

Success response

{
  "Success": true,
  "Action": "LoadProfile",
  "Message": "Profile load requested",
  "ProfileGuid": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c"
}

Errors

MISSING_PARAMETER, neither ProfileGuid nor ProfileName was provided.
INTERNAL_ERROR, unexpected exception while loading.

NOTE: If the supplied ProfileGuid / ProfileName doesn't resolve, the server still returns Success: true, "Profile load requested" and silently does nothing. Validate references with GetProfiles first if you need to detect bad input.

SwapToVoiceChangerMode

Added in 1.0.4.5

Switch the main UI to voice-changer mode.

Success response

{
  "Success": true,
  "Action": "SwapToVoiceChangerMode",
  "Message": "Switched to voice changer mode"
}

SwapToSoundboardMode

Added in 1.0.4.5

Switch the main UI to soundboard mode.

Success response

{
  "Success": true,
  "Action": "SwapToSoundboardMode",
  "Message": "Switched to soundboard mode"
}

Legacy

These older profile-switching actions remain for back-compat. Prefer LoadProfile for new integrations.

OpenProfilePage

Added in 1.0.4.5

Switch to a profile and load its UI page.

Request fields

ProfileGuid (preferred) or ProfileName, required.

Example request

{
  "Action": "OpenProfilePage",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "ProfileGuid": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c"
}

Success responses

OpenProfilePage emits two messages back-to-back: a detailed OpenProfilePage payload with the profile image and full SFX-button data, followed by a brief ProfileLoaded confirmation. Both arrive on the same socket; if you set a RequestId on the request, both responses echo it back, so match on Action to tell them apart.

// 1. OpenProfilePage response (full payload)
{
  "Success": true,
  "Action": "OpenProfilePage",
  "Message": "Profile 'Gaming' is now active.",
  "ProfileImagePath": "C:\path\to\profile-image.png",
  "SfxButtons": [ /* all button data for the profile */ ],
  "RequestId": "<echoed if you sent one>"
}

// 2. ProfileLoaded response (confirmation)
{
  "Success": true,
  "Action": "ProfileLoaded",
  "ProfileName": "Gaming",
  "ProfileGuid": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c",
  "RequestId": "<echoed if you sent one>"
}

NOTE: If you only need confirmation that the profile switched, branch on the second response's Action: "ProfileLoaded". If you need the image and button data, capture the first one.

SetProfile

Added in 1.0.4.5

Set the active profile (similar to OpenProfilePage, kept for back-compat). Sends a single response, no follow-up ProfileLoaded message.

Request fields

ProfileGuid (preferred) or ProfileName, required.

Success response

{
  "Success": true,
  "Action": "SetProfile",
  "Message": "Profile 'Gaming' has been set successfully.",
  "ProfileName": "Gaming",
  "ProfileGuid": "8f3a1c20-5c7e-4b6e-9b3d-1d2e4f5a6b7c",
  "ProfileImagePath": "C:\path\to\profile-image.png"
}

Errors

PROFILE_NOT_FOUND, the supplied ProfileGuid / ProfileName doesn't resolve.
INTERNAL_ERROR, unexpected exception while loading the profile.

Microphone

SmoredBoard's mic has two independent layers: the loopback push-to-talk mute (a fast soft-mute that gates audio without tearing down the audio pipeline) and main mic on/off (a hard teardown that fully stops and restarts the loopback chain). The mute layer is what you want for fast or frequent toggling, use the on/off layer only when you genuinely want to start or stop the entire pipeline.

NOTE: All mic actions below (MuteMic, UnmuteMic, ToggleMuteMic, ToggleMic, StartMic, StopMic) return Success: true even when the loopback audio pipeline isn't initialized yet, in which case they silently no-op. If you need to verify the mic actually changed state, follow up with GetMicMuteState or GetMicState.

MuteMic

Added in 1.0.4.5

Mute the loopback push-to-talk layer. Same operation as the in-app PTT keybind.

Success response

{
  "Success": true,
  "Action": "MuteMic",
  "Message": "Mic muted"
}

UnmuteMic

Added in 1.0.4.5

Unmute the loopback push-to-talk layer.

Success response

{
  "Success": true,
  "Action": "UnmuteMic",
  "Message": "Mic unmuted"
}

ToggleMuteMic

Added in 1.0.4.5

Toggle the loopback push-to-talk state. Same operation as the in-app PTT keybind.

Success response

{
  "Success": true,
  "Action": "ToggleMuteMic",
  "Message": "Mic mute toggled"
}

GetMicMuteState

Added in 1.0.4.5

Read the loopback push-to-talk mute state (separate from the main mic on/off).

Success response

{
  "Success": true,
  "Action": "MicMuteStateUpdate",
  "Message": "unmuted"
}

Message is "muted" or "unmuted".

NOTE: The response Action is "MicMuteStateUpdate", not "GetMicMuteState".

Main mic on/off:

NOTE: These actions tear down or rebuild the entire loopback chain, so turning the mic off and back on is noticeably less smooth than muting it. If you just need to mute frequently or quickly, use MuteMic / UnmuteMic / ToggleMuteMic above instead.

ToggleMic

Added in 1.0.4.5

Toggle the main microphone on/off.

Success response

{
  "Success": true,
  "Action": "ToggleMic",
  "Message": "on"
}

Message is "on" or "off", reflecting the new state.

GetMicState

Added in 1.0.4.5

Read whether the main mic is currently on or off.

Success response

{
  "Success": true,
  "Action": "MicStateUpdate",
  "Message": "on"
}

Message is "on" or "off".

NOTE: The response Action is "MicStateUpdate", not "GetMicState".

StartMic

Added in 1.0.4.5

Turn the mic on if it's off. No-op if already on.

Success response

{
  "Success": true,
  "Action": "StartMic",
  "Message": "Mic started"
}

StopMic

Added in 1.0.4.5

Turn the mic off if it's on. No-op if already off.

Success response

{
  "Success": true,
  "Action": "StopMic",
  "Message": "Mic stopped"
}

Voice Changer

GetAllVoiceChangers

Added in 1.0.4.5

List every voice changer available, built-in, custom, plus a None entry.

Example request

{
  "Action": "GetAllVoiceChangers",
  "Token": "SMORED1999VERYGOODANDCOOL"
}

Success response

{
  "Success": true,
  "Action": "GetAllVoiceChangers",
  "VoiceChangers": [
    { "id": "__NONE__", "name": "None", "IsBuiltIn": false },
    { "id": "builtin:robot1", "name": "Robot", "IsBuiltIn": true },
    { "id": "1be13752690e4883a4ab92a642aff588", "name": "My Custom VC", "IsBuiltIn": false }
  ]
}

Use the id field as VoiceChangerId for SetVoiceChangerById or GetVoiceChangerDetails. The "__NONE__" entry is equivalent to TurnOffVoiceChanger.

GetVoiceChangerDetails

Added in 1.0.4.5

Fetch metadata and image data for a single voice changer by ID. Use this when you need the icon or display info for a specific voice changer (e.g. to render a button preview).

Request fields

VoiceChangerId (string, preferred), the built-in or custom voice changer ID. Use "__NONE__" for the no-voice-changer entry.
ProfileName (string, legacy fallback), accepted for legacy callers, prefer VoiceChangerId in new code.

Example request

{
  "Action": "GetVoiceChangerDetails",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "VoiceChangerId": "builtin:robot1"
}

Success response

{
  "Success": true,
  "Action": "GetVoiceChangerDetails",
  "VoiceChangerId": "builtin:robot1",
  "VoiceChangerName": "Robot",
  "ImageData": "<base64 PNG or empty string>",
  "ImageType": "base64"
}

ImageType is "base64" when ImageData contains a base64-encoded PNG, or "none" when no image is available.

Errors

MISSING_PARAMETER, no VoiceChangerId (or legacy ProfileName) provided.
VOICE_CHANGER_NOT_FOUND, the supplied id doesn't match any built-in or custom voice changer.
INTERNAL_ERROR, unexpected server error during lookup.

NOTE: This action does not include an isBuiltIn flag in the response. If you need to know whether an id is built-in or a user-created custom voice changer, cross-reference against GetAllVoiceChangers.

SetVoiceChangerById

Added in 1.0.4.5

Activate a voice changer by its ID.

Request fields

VoiceChangerId (string, required), built-in or custom voice changer ID. Use "__NONE__" to clear.

Example request

{
  "Action": "SetVoiceChangerById",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "VoiceChangerId": "builtin:robot1"
}

Success response

{
  "Success": true,
  "Action": "SetVoiceChangerById",
  "Message": "Voice changer set"
}

Errors

MISSING_PARAMETER, no VoiceChangerId provided.

NOTE: Get available voice changer IDs by calling GetAllVoiceChangers. If the supplied VoiceChangerId doesn't match any built-in or custom voice changer, the server still returns Success: true, "Voice changer set" and silently does nothing. Validate IDs with GetAllVoiceChangers if you need to detect bad input.

TurnOffVoiceChanger

Added in 1.0.4.5

Disable the active voice changer.

Success response

{
  "Success": true,
  "Action": "TurnOffVoiceChanger",
  "Message": "Voice changer turned off"
}

Censor

CensorOn

Added in 1.0.4.5

Start the censor beep (until turned off).

Success response

{
  "Success": true,
  "Action": "CensorOn",
  "Message": "Censor on"
}

CensorOff

Added in 1.0.4.5

Stop the censor beep.

Success response

{
  "Success": true,
  "Action": "CensorOff",
  "Message": "Censor off"
}

ToggleCensor

Added in 1.0.4.5

Toggle the censor beep on/off.

Success response

{
  "Success": true,
  "Action": "ToggleCensor",
  "Message": "Censor toggled"
}

CensorForTime

Added in 1.0.4.5

Turn censor on for a fixed duration, then auto-revert to its prior state.

Request fields

Seconds (number, required, > 0), duration. Capped at 24 hours server-side.

Example request

{
  "Action": "CensorForTime",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "Seconds": 3.5
}

Success response

{
  "Success": true,
  "Action": "CensorForTime",
  "Message": "Censor on for 3.5s"
}

Errors

MISSING_PARAMETER, no Seconds provided.
INVALID_PARAMETER, Seconds is zero or negative.

Sound Snap

GetSoundSnapState

Added in 1.0.4.5

Read whether Sound Snap is on.

Success response

{
  "Success": true,
  "Action": "SoundSnapStateUpdate",
  "Message": "on"
}

NOTE: The response Action is "SoundSnapStateUpdate", not "GetSoundSnapState".

Errors

NOT_AVAILABLE, the main window isn't ready (rare, briefly during app startup). Message is "MainWindow not available". Treat as transient and retry.

SoundSnapOnPro

Added in 1.0.4.5

Enable Sound Snap. No-op if already on.

Success response

{
  "Success": true,
  "Action": "SoundSnapOn",
  "Message": "Sound snap on"
}

SoundSnapOffPro

Added in 1.0.4.5

Disable Sound Snap. No-op if already off.

Success response

{
  "Success": true,
  "Action": "SoundSnapOff",
  "Message": "Sound snap off"
}

ToggleSoundSnapPro

Added in 1.0.4.5

Toggle Sound Snap on/off.

Success response

{
  "Success": true,
  "Action": "ToggleSoundSnap",
  "Message": "Sound snap toggled"
}

SoundSnapForTimePro

Added in 1.0.4.5

Enable Sound Snap for a duration, then revert to its prior state.

Request fields

Seconds (number, required, > 0), duration. Capped at 24 hours.

Example request

{
  "Action": "SoundSnapForTime",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "Seconds": 10
}

Success response

{
  "Success": true,
  "Action": "SoundSnapForTime",
  "Message": "Sound snap on for 10s"
}

SoundSnapPro

Added in 1.0.4.5

Fire the in-app Sound Snap button (the toolbar button that toggles Sound Snap). Equivalent to a user clicking the button in the UI. Takes no extra fields.

Success response

{
  "Success": true,
  "Action": "SoundSnap"
}

Errors

PRO_REQUIRED, the current user isn't a Pro subscriber. Message is "Pro required".
NOT_AVAILABLE, the main window isn't ready (rare, briefly during app startup). Message is "MainWindow not available".

NOTE: For deterministic on/off/toggle control, prefer SoundSnapOn, SoundSnapOff, or ToggleSoundSnap. SoundSnap exists for UI-mirror integrations that just want to fire the button.

Random Sound Mode

GetRandomSoundModeStatus

Added in 1.0.4.5

Read whether random-sound mode is on.

Success response

{
  "Success": true,
  "Action": "RandomSoundModeStateUpdate",
  "Message": "on"
}

NOTE: The response Action is "RandomSoundModeStateUpdate", not "GetRandomSoundModeStatus".

RandomSoundModeOnPro

Added in 1.0.4.5

Enable random-sound mode. No-op if already on.

Success response

{
  "Success": true,
  "Action": "RandomSoundModeOn",
  "Message": "Random sound mode on"
}

RandomSoundModeOffPro

Added in 1.0.4.5

Disable random-sound mode. No-op if already off.

Success response

{
  "Success": true,
  "Action": "RandomSoundModeOff",
  "Message": "Random sound mode off"
}

RandomSoundModeTogglePro

Added in 1.0.4.5

Toggle random-sound mode.

Success response

{
  "Success": true,
  "Action": "RandomSoundModeToggle",
  "Message": "Random sound mode toggled"
}

RandomSoundModePro

Added in 1.0.4.5

Fire the in-app Random Sound Mode button (the toolbar control that toggles random-sound mode). Equivalent to a user clicking the button in the UI. Takes no extra fields.

Success response

{
  "Success": true,
  "Action": "RandomSoundMode"
}

Errors

PRO_REQUIRED, the current user isn't a Pro subscriber. Message is "Pro required".

NOTE: For deterministic on/off/toggle control, prefer RandomSoundModeOn, RandomSoundModeOff, or RandomSoundModeToggle. RandomSoundMode exists for UI-mirror integrations that just want to fire the button.

SFX Output

GetSfxOutputState

Added in 1.0.4.5

Read whether SFX output is on.

Success response

{
  "Success": true,
  "Action": "SfxOutputStateUpdate",
  "Message": "on"
}

NOTE: The response Action is "SfxOutputStateUpdate", not "GetSfxOutputState".

ToggleSfxOutput

Added in 1.0.4.5

Toggle SFX output on/off.

Success response

{
  "Success": true,
  "Action": "SfxOutputStateUpdate",
  "Message": "on"
}

Message is the new state.

NOTE: The response Action is "SfxOutputStateUpdate", not "ToggleSfxOutput" (toggle reuses the state-update channel).

Hear Self

GetHearSelfState

Added in 1.0.4.5

Read whether self-monitoring is on.

Success response

{
  "Success": true,
  "Action": "HearSelfStateUpdate",
  "Message": "on"
}

NOTE: The response Action is "HearSelfStateUpdate", not "GetHearSelfState".

ToggleHearSelf

Added in 1.0.4.5

Toggle self-monitoring on/off.

Success response

{
  "Success": true,
  "Action": "ToggleHearSelf",
  "Message": "on"
}

Message is the new state.

Keybinds

GetKeybindState

Added in 1.0.4.5

Read whether the global keybind system is enabled.

Success response

{
  "Success": true,
  "Action": "KeybindStateUpdate",
  "Message": "on"
}

NOTE: The response Action is "KeybindStateUpdate", not "GetKeybindState".

ToggleKeybind

Added in 1.0.4.5

Toggle the global keybind system on/off.

Success response

{
  "Success": true,
  "Action": "ToggleKeybind",
  "Message": "on"
}

Message is the new state.

Automations

TriggerAutomationPro

Added in 1.0.4.5

Fire an automation by its ID. Same as the automation firing from any other source (keybind, chat trigger, schedule).

Request fields

AutomationId (string, required), GUID of the automation to fire.

Example request

{
  "Action": "TriggerAutomation",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "AutomationId": "a1b2c3d4-e5f6-7890-abcd-ef1234567890"
}

Success response

{
  "Success": true,
  "Action": "TriggerAutomation",
  "Message": "Automation triggered"
}

Errors

MISSING_PARAMETER, no AutomationId provided.
AUTOMATION_NOT_FOUND, the supplied AutomationId doesn't match any existing automation.
PRO_REQUIRED, the current user isn't a Pro subscriber.

Chat Triggers

Pause, resume, or toggle the chat triggers that fire when messages arrive in your connected Twitch / YouTube chats. Two scopes are supported: a single global pause that affects every connected stream, and a per-stream pause that affects one configured chat instance.

The *ForTime commands pause for a duration and then auto-resume, with one exception: if triggers were already paused when the timed pause started, they stay paused after the timer expires (the timer never overrides a pre-existing pause). Use GetAllChatStreams to look up valid StreamId values for the per-stream commands.

NOTE: All 8 mutation commands below are also available as first-class entries in chat triggers and automations under a new "Chat Triggers" category, so you can alternatively wrap them in an automation and invoke them via TriggerAutomation.

GetGlobalChatTriggersState

Added in 1.0.4.5

Read whether the global chat-trigger pause is currently active. Cheap, no side effects. Takes no extra fields.

Success response

{
  "Success": true,
  "Action": "GetGlobalChatTriggersState",
  "TriggersPaused": false
}

NOTE: This only reports the global pause flag. To inspect per-stream pause state, call GetAllChatStreams and read the TriggersPaused field on each returned stream.

GetAllChatStreams

Added in 1.0.4.5

List every configured Twitch / YouTube chat stream, including each stream's current per-stream pause state. Use the returned Id as StreamId in subsequent per-stream commands.

Success response

{
  "Success": true,
  "Action": "GetAllChatStreams",
  "ChatStreams": [
    {
      "Id": "550e8400-e29b-41d4-a716-446655440000",
      "DisplayName": "MyTwitchChannel",
      "Platform": "Twitch",
      "Identifier": "mytwitchchannel",
      "IsEnabled": true,
      "TriggersPaused": false
    }
  ]
}

Per-stream fields

Id (string), GUID of the chat instance. Stable across renames. Pass this as StreamId in per-stream commands.
DisplayName (string), human-friendly name shown in the in-app chat configurator.
Platform (string), one of "Twitch" or "YouTube".
Identifier (string), platform-specific identifier (Twitch login, YouTube channel ID, etc.).
IsEnabled (boolean), whether the stream is enabled in the chat configurator.
TriggersPaused (boolean), whether per-stream pause is currently active for this stream.

Errors

INTERNAL_ERROR, the server caught an unexpected exception while reading chat settings.

NOTE: When zero streams are configured, ChatStreams is omitted from the response entirely (not returned as []). Default missing-field handling to an empty list on the client.

PauseAllChatTriggers

Added in 1.0.4.5

Pause every chat trigger globally, across every connected stream. No-op if global pause is already on. Takes no extra fields.

Success response

{
  "Success": true,
  "Action": "PauseAllChatTriggers",
  "Message": "All chat triggers paused",
  "TriggersPaused": true
}

NOTE: Global pause is in-memory only and resets on app restart. For persistent per-stream pause, use PauseStreamTriggers.

ResumeAllChatTriggers

Added in 1.0.4.5

Resume chat triggers globally. No-op if global pause was already off. Does not affect per-stream pause state, a stream paused via PauseStreamTriggers stays paused. Takes no extra fields.

Success response

{
  "Success": true,
  "Action": "ResumeAllChatTriggers",
  "Message": "All chat triggers resumed",
  "TriggersPaused": false
}

ToggleAllChatTriggers

Added in 1.0.4.5

Flip the global chat-trigger pause flag. Takes no extra fields.

Success response

{
  "Success": true,
  "Action": "ToggleAllChatTriggers",
  "Message": "All chat triggers paused",
  "TriggersPaused": true
}

Message reflects the resulting state ("All chat triggers paused" or "All chat triggers resumed"); TriggersPaused matches.

PauseAllChatTriggersForTime

Added in 1.0.4.5

Pause chat triggers globally for a duration, then auto-resume. If global triggers were already paused when this call started, they stay paused after the timer expires (the timer never overrides a pre-existing pause).

Request fields

Seconds (number, required, > 0), duration. Fractional values are allowed. Capped at 24 hours.

Example request

{
  "Action": "PauseAllChatTriggersForTime",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "Seconds": 30
}

Success response

{
  "Success": true,
  "Action": "PauseAllChatTriggersForTime",
  "Message": "All chat triggers paused for 30s",
  "TriggersPaused": true
}

Errors

MISSING_PARAMETER, Seconds is missing or not greater than 0. Message is "Seconds (> 0) is required".

NOTE: Overlapping calls are safe. Each new call restarts the timer with its own duration; only the most recent call's scheduled resume can fire.

PauseStreamTriggers

Added in 1.0.4.5

Pause chat triggers for a single configured chat stream. The stream stays paused until you resume or toggle it. State is persisted to chatSettings.json and survives app restarts.

Request fields

StreamId (string, required), GUID of a configured chat stream from GetAllChatStreams.

Example request

{
  "Action": "PauseStreamTriggers",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "StreamId": "550e8400-e29b-41d4-a716-446655440000"
}

Success response

{
  "Success": true,
  "Action": "PauseStreamTriggers",
  "Message": "Stream triggers paused",
  "StreamId": "550e8400-e29b-41d4-a716-446655440000",
  "TriggersPaused": true
}

Errors

MISSING_PARAMETER, no StreamId provided.
STREAM_NOT_FOUND, the supplied StreamId doesn't match any configured chat stream.

ResumeStreamTriggers

Added in 1.0.4.5

Resume chat triggers for a single configured chat stream. State is persisted to chatSettings.json.

Request fields

StreamId (string, required).

Success response

{
  "Success": true,
  "Action": "ResumeStreamTriggers",
  "Message": "Stream triggers resumed",
  "StreamId": "550e8400-e29b-41d4-a716-446655440000",
  "TriggersPaused": false
}

Errors

MISSING_PARAMETER, no StreamId provided.
STREAM_NOT_FOUND, the supplied StreamId doesn't match any configured chat stream.

ToggleStreamTriggers

Added in 1.0.4.5

Flip the per-stream pause flag for a single configured chat stream. State is persisted to chatSettings.json.

Request fields

StreamId (string, required).

Success response

{
  "Success": true,
  "Action": "ToggleStreamTriggers",
  "Message": "Stream triggers toggled",
  "StreamId": "550e8400-e29b-41d4-a716-446655440000",
  "TriggersPaused": true
}

TriggersPaused reflects the actual stored state after the toggle. Branch on it rather than inferring from the request.

Errors

MISSING_PARAMETER, no StreamId provided.
STREAM_NOT_FOUND, the supplied StreamId doesn't match any configured chat stream.

PauseStreamTriggersForTime

Added in 1.0.4.5

Pause a single configured chat stream for a duration, then auto-resume. If the stream was already paused when this call started, it stays paused after the timer expires.

Request fields

StreamId (string, required).
Seconds (number, required, > 0), duration. Fractional values are allowed. Capped at 24 hours.

Example request

{
  "Action": "PauseStreamTriggersForTime",
  "Token": "SMORED1999VERYGOODANDCOOL",
  "StreamId": "550e8400-e29b-41d4-a716-446655440000",
  "Seconds": 30
}

Success response

{
  "Success": true,
  "Action": "PauseStreamTriggersForTime",
  "Message": "Stream triggers paused for 30s",
  "StreamId": "550e8400-e29b-41d4-a716-446655440000",
  "TriggersPaused": true
}

Errors

MISSING_PARAMETER, no StreamId provided, or Seconds is missing or not greater than 0. Message is either "StreamId is required" or "Seconds (> 0) is required".
STREAM_NOT_FOUND, the supplied StreamId doesn't match any configured chat stream.

NOTE: Overlapping calls per stream are safe. Each new call restarts that stream's timer with its own duration.

Diagnostics

Hello

Added in 1.0.4.5

Lightweight ping to confirm the WebSocket server is reachable and your Token is valid. Useful for health checks, connection probes, and quick manual sanity tests. Takes no extra fields.

Example request

{
  "Action": "Hello",
  "Token": "SMORED1999VERYGOODANDCOOL"
}

Success response

{
  "Success": true,
  "Action": "Hello",
  "Message": "Hello! The SmoredBoard WebSocket server is reachable!"
}

NOTE: Hello accepts a lowercase alias "hello" for ad-hoc CLI typing. Use the canonical PascalCase form "Hello" in code. Most other actions are case-sensitive, so don't rely on this pattern elsewhere.

ResetWindow

Added in 1.0.4.5

Reset the SmoredBoard main window to its default size and position. Useful when the window has been dragged off-screen or onto a disconnected monitor. Takes no extra fields.

Success response

{
  "Success": true,
  "Action": "ResetWindow"
}

Errors

NOT_AVAILABLE, the main window isn't ready (rare, briefly during app startup). Message is "MainWindow not available".

Examples

A common pattern is to discover everything dynamically instead of hardcoding paths and GUIDs. Start with GetProfiles to list profiles, use the returned Id as ProfileGuid to fetch sounds via GetProfileSounds, then trigger one with PlaySound. The example below chains those three calls in a single browser script.

const ws = new WebSocket('ws://127.0.0.1:8181');
const TOKEN = 'SMORED1999VERYGOODANDCOOL';

let profileGuid = null;

ws.addEventListener('open', () => {
  // Step 1: list available profiles.
  ws.send(JSON.stringify({ Action: 'GetProfiles', Token: TOKEN }));
});

ws.addEventListener('message', (event) => {
  const data = JSON.parse(event.data);
  if (!data.Success) {
    console.error('Error:', data.Message);
    return;
  }

  if (data.Action === 'GetProfiles') {
    // Step 2: pick a profile (here, the first one) and ask for its sounds.
    const firstProfile = data.Profiles[0];
    profileGuid = firstProfile.Id;
    console.log('Picked profile:', firstProfile.name);

    ws.send(JSON.stringify({
      Action: 'GetProfileSounds',
      Token: TOKEN,
      ProfileGuid: profileGuid,
    }));
  } else if (data.Action === 'GetProfileSounds') {
    // Step 3: pick a sound (here, the first one) and play it.
    const firstSound = data.Sounds[0];
    if (!firstSound) return;
    console.log('Playing:', firstSound.Name);

    ws.send(JSON.stringify({
      Action: 'PlaySound',
      Token: TOKEN,
      SoundPath: firstSound.SoundPath,
      ProfileGuid: profileGuid,
    }));
  } else if (data.Action === 'PlaySound') {
    console.log('Sound triggered.');
  }
});