Files
openclaw/extensions/xai/capability-provider-metadata.ts
Peter Steinberger fa03d9b913 refactor: consolidate coercion helpers (#121366)
* refactor: consolidate coercion helpers

* fix: remove duplicate coercion imports

* fix: preserve serialized coercion guard

* chore: ratchet coercion helper carve-outs

* fix(test): keep gauntlet subprocess startup lean

* fix: preserve imported session timestamp semantics

* fix: preserve catalog timestamp string semantics

* chore: align plugin SDK surface ratchet

* fix: preserve trajectory and SDK string contracts

* fix(test): preserve QA record assertion semantics

* fix: complete standalone record guard rename

* refactor(cron): use canonical string coercion

* fix(acpx): preserve Pi timestamp parsing

* test(channels): adapt custody test harnesses

* test(telegram): classify media harness as test support

* test(acpx): split timestamp contract coverage

* test(channels): support generated custody contracts

* chore: ban the full coercion helper name set

Extends the declaration guard to all eleven consolidated helper names and
renames the cron schedule-identity readNumber wrapper to readScheduleInteger
so the banned generic name cannot regrow.

* fix(scripts): repair release-validation guard drift and lint cause

Restores the renamed isJsonRecord guard in assertTrustedWorkflowHarness after
main added isRecord call sites in parallel, and attaches the caught YAML error
as the thrown error cause (preserve-caught-error was red on main).

* fix: preserve Claude timestamp string semantics

* fix: preserve persisted timestamp string semantics

* fix: preserve date-first timestamp contracts

* fix(openai): harden delegation failure formatting

* chore: close coercion helper guard gaps

* test(openai): model non-error delegation rejection

* chore: refresh plugin SDK API contract

* fix(tasks): use canonical string field reader

* fix(ai): use canonical provider error field coercion

* fix(browser): migrate native bootstrap coercion

* docs(plugin-sdk): clarify text record export compatibility

* fix(gateway): normalize approval execution identity

* test(outbound): isolate message action poll harness
2026-08-11 00:02:18 -07:00

289 lines
9.2 KiB
TypeScript

import type { ImageGenerationProvider } from "openclaw/plugin-sdk/image-generation";
import type { MediaUnderstandingProvider } from "openclaw/plugin-sdk/media-understanding";
import {
isProviderApiKeyConfigured,
isProviderAuthProfileConfigured,
} from "openclaw/plugin-sdk/provider-auth";
import type {
RealtimeTranscriptionProviderConfig,
RealtimeTranscriptionProviderPlugin,
} from "openclaw/plugin-sdk/realtime-transcription";
import type {
RealtimeVoiceAudioFormat,
RealtimeVoiceBridgeCreateRequest,
RealtimeVoiceProviderPlugin,
} from "openclaw/plugin-sdk/realtime-voice";
import { normalizeResolvedSecretInputString } from "openclaw/plugin-sdk/secret-input";
import {
isRecord,
normalizeOptionalString,
parseBooleanValue,
parseFiniteNumber,
} from "openclaw/plugin-sdk/string-coerce-runtime";
import type {
VideoGenerationProvider,
VideoGenerationProviderCapabilities,
} from "openclaw/plugin-sdk/video-generation";
import { XAI_DEFAULT_IMAGE_MODEL, XAI_IMAGE_MODELS } from "./model-definitions.js";
import {
XAI_REALTIME_DEFAULT_MODEL,
XAI_REALTIME_VOICES,
hasXaiRealtimeApiKeyInput,
normalizeXaiRealtimeProviderConfig,
} from "./realtime-voice-config.js";
export const XAI_IMAGE_DEFAULT_TIMEOUT_MS = 600_000;
export const XAI_SUPPORTED_IMAGE_ASPECT_RATIOS = [
"1:1",
"16:9",
"9:16",
"4:3",
"3:4",
"3:2",
"2:3",
"2:1",
"1:2",
"19.5:9",
"9:19.5",
"20:9",
"9:20",
] as const;
export function createXaiImageGenerationProviderMetadata() {
return {
id: "xai",
label: "xAI",
defaultModel: XAI_DEFAULT_IMAGE_MODEL,
defaultTimeoutMs: XAI_IMAGE_DEFAULT_TIMEOUT_MS,
models: [...XAI_IMAGE_MODELS],
capabilities: {
generate: {
maxCount: 4,
supportsAspectRatio: true,
supportsResolution: true,
supportsSize: false,
},
edit: {
enabled: true,
maxCount: 4,
maxInputImages: 3,
supportsAspectRatio: true,
supportsResolution: true,
supportsSize: false,
},
geometry: {
aspectRatios: [...XAI_SUPPORTED_IMAGE_ASPECT_RATIOS],
resolutions: ["1K", "2K"],
},
},
} satisfies Omit<ImageGenerationProvider, "generateImage">;
}
export function createXaiMediaUnderstandingProviderMetadata() {
return {
id: "xai",
capabilities: ["audio"],
autoPriority: { audio: 25 },
} satisfies Omit<MediaUnderstandingProvider, "transcribeAudio">;
}
export const DEFAULT_XAI_VIDEO_BASE_URL = "https://api.x.ai/v1";
export const DEFAULT_XAI_VIDEO_MODEL = "grok-imagine-video";
const XAI_VIDEO_15_MODEL = "grok-imagine-video-1.5";
export const XAI_VIDEO_DEFAULT_TIMEOUT_MS = 600_000;
export const XAI_VIDEO_ASPECT_RATIOS = new Set(["1:1", "16:9", "9:16", "4:3", "3:4", "3:2", "2:3"]);
const XAI_VIDEO_15_CAPABILITIES = {
imageToVideo: {
enabled: true,
maxVideos: 1,
maxInputImages: 1,
maxDurationSeconds: 15,
aspectRatios: [...XAI_VIDEO_ASPECT_RATIOS],
resolutions: ["480P", "720P", "1080P"],
supportsAspectRatio: true,
supportsResolution: true,
},
videoToVideo: {
enabled: false,
},
} satisfies VideoGenerationProviderCapabilities;
const XAI_VIDEO_15_MODEL_IDS = new Set([
XAI_VIDEO_15_MODEL,
"grok-imagine-video-1.5-preview",
"grok-imagine-video-1.5-2026-05-30",
]);
export function isXaiVideo15Model(model: string | undefined): boolean {
const normalized = normalizeOptionalString(model);
return normalized ? XAI_VIDEO_15_MODEL_IDS.has(normalized) : false;
}
export function createXaiVideoGenerationProviderMetadata() {
return {
id: "xai",
label: "xAI",
defaultModel: DEFAULT_XAI_VIDEO_MODEL,
defaultTimeoutMs: XAI_VIDEO_DEFAULT_TIMEOUT_MS,
models: [DEFAULT_XAI_VIDEO_MODEL, XAI_VIDEO_15_MODEL],
catalogByModel: {
[XAI_VIDEO_15_MODEL]: {
capabilities: XAI_VIDEO_15_CAPABILITIES,
modes: ["imageToVideo"],
},
},
isConfigured: (ctx) => isProviderApiKeyConfigured({ provider: "xai", ...ctx }),
capabilities: {
generate: {
maxVideos: 1,
maxDurationSeconds: 15,
aspectRatios: [...XAI_VIDEO_ASPECT_RATIOS],
resolutions: ["480P", "720P"],
supportsAspectRatio: true,
supportsResolution: true,
},
imageToVideo: {
enabled: true,
maxVideos: 1,
maxInputImages: 7,
maxDurationSeconds: 15,
aspectRatios: [...XAI_VIDEO_ASPECT_RATIOS],
resolutions: ["480P", "720P"],
supportsAspectRatio: true,
supportsResolution: true,
},
videoToVideo: {
enabled: true,
maxVideos: 1,
maxInputVideos: 1,
maxDurationSeconds: 10,
supportsAspectRatio: false,
supportsResolution: false,
},
},
resolveModelCapabilities: ({ model }) =>
isXaiVideo15Model(model) ? XAI_VIDEO_15_CAPABILITIES : undefined,
} satisfies Omit<VideoGenerationProvider, "generateVideo">;
}
export type XaiRealtimeTranscriptionEncoding = "pcm" | "mulaw" | "alaw";
type XaiRealtimeTranscriptionProviderConfig = {
apiKey?: string;
baseUrl?: string;
sampleRate?: number;
encoding?: XaiRealtimeTranscriptionEncoding;
interimResults?: boolean;
endpointingMs?: number;
language?: string;
};
function normalizeRealtimeTranscriptionEncoding(
value: unknown,
): XaiRealtimeTranscriptionEncoding | undefined {
const normalized = normalizeOptionalString(value)?.toLowerCase();
if (!normalized) {
return undefined;
}
if (normalized === "ulaw" || normalized === "g711_ulaw" || normalized === "g711-mulaw") {
return "mulaw";
}
if (normalized === "g711_alaw" || normalized === "g711-alaw") {
return "alaw";
}
if (normalized === "pcm" || normalized === "mulaw" || normalized === "alaw") {
return normalized;
}
throw new Error(`Invalid xAI realtime transcription encoding: ${normalized}`);
}
export function normalizeXaiRealtimeTranscriptionProviderConfig(
config: RealtimeTranscriptionProviderConfig,
): XaiRealtimeTranscriptionProviderConfig {
const raw = isRecord(config) ? config : undefined;
const providers = isRecord(raw?.providers) ? raw.providers : undefined;
const nested = providers?.xai ?? raw?.xai ?? raw;
const xai = isRecord(nested) ? nested : {};
return {
apiKey: normalizeResolvedSecretInputString({
value: xai.apiKey,
path: "plugins.entries.voice-call.config.streaming.providers.xai.apiKey",
}),
baseUrl: normalizeOptionalString(xai.baseUrl),
sampleRate: parseFiniteNumber(xai.sampleRate ?? xai.sample_rate),
encoding: normalizeRealtimeTranscriptionEncoding(xai.encoding),
interimResults: parseBooleanValue(xai.interimResults ?? xai.interim_results),
endpointingMs: parseFiniteNumber(xai.endpointingMs ?? xai.endpointing ?? xai.silenceDurationMs),
language: normalizeOptionalString(xai.language),
};
}
export function createXaiRealtimeTranscriptionProviderMetadata() {
return {
id: "xai",
label: "xAI Realtime Transcription",
aliases: ["xai-realtime", "grok-stt-streaming"],
autoSelectOrder: 25,
resolveConfig: ({ rawConfig }) => normalizeXaiRealtimeTranscriptionProviderConfig(rawConfig),
isConfigured: ({ providerConfig, cfg }) =>
Boolean(
normalizeXaiRealtimeTranscriptionProviderConfig(providerConfig).apiKey ??
normalizeOptionalString(process.env.XAI_API_KEY),
) || isProviderAuthProfileConfigured({ provider: "xai", cfg }),
} satisfies Omit<RealtimeTranscriptionProviderPlugin, "createSession">;
}
const XAI_REALTIME_AUDIO_FORMAT_G711_ULAW_8KHZ = {
encoding: "g711_ulaw",
sampleRateHz: 8000,
channels: 1,
} satisfies RealtimeVoiceAudioFormat;
const XAI_REALTIME_AUDIO_FORMAT_PCM16_24KHZ = {
encoding: "pcm16",
sampleRateHz: 24000,
channels: 1,
} satisfies RealtimeVoiceAudioFormat;
export function createXaiRealtimeVoiceProviderMetadata() {
return {
id: "xai",
label: "xAI Grok Voice",
aliases: ["xai-realtime-voice", "grok-voice"],
defaultModel: XAI_REALTIME_DEFAULT_MODEL,
voices: XAI_REALTIME_VOICES,
autoSelectOrder: 25,
capabilities: {
transports: ["gateway-relay"],
inputAudioFormats: [
XAI_REALTIME_AUDIO_FORMAT_G711_ULAW_8KHZ,
XAI_REALTIME_AUDIO_FORMAT_PCM16_24KHZ,
],
outputAudioFormats: [
XAI_REALTIME_AUDIO_FORMAT_G711_ULAW_8KHZ,
XAI_REALTIME_AUDIO_FORMAT_PCM16_24KHZ,
],
supportsBargeIn: true,
handlesInputAudioBargeIn: true,
supportsToolCalls: true,
supportsSessionResumption: true,
},
resolveConfig: ({ rawConfig }) => normalizeXaiRealtimeProviderConfig(rawConfig),
isConfigured: ({ providerConfig, cfg }) =>
hasXaiRealtimeApiKeyInput(normalizeXaiRealtimeProviderConfig(providerConfig).apiKey, cfg),
} satisfies Omit<RealtimeVoiceProviderPlugin, "createBridge" | "createBrowserSession">;
}
export function assertXaiRealtimeVoiceRequestSupported(
req: RealtimeVoiceBridgeCreateRequest,
): void {
const config = normalizeXaiRealtimeProviderConfig(req.providerConfig);
if (req.autoRespondToAudio === false) {
throw new Error(
'xAI realtime voice requires automatic server-VAD responses; use consultRouting: "provider-direct"',
);
}
if ((req.interruptResponseOnInputAudio ?? config.interruptResponseOnInputAudio) === false) {
throw new Error("xAI realtime voice requires automatic server-VAD interruption handling");
}
}