@midscene/core 0.30.2-beta-20251010092125.0 → 0.30.3-beta-20251011064436.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -68,12 +68,16 @@ export declare class Agent<InterfaceType extends AbstractInterface = AbstractInt
68
68
  value: string;
69
69
  } & {
70
70
  autoDismissKeyboard?: boolean;
71
+ } & {
72
+ append?: boolean;
71
73
  }): Promise<any>;
72
74
  /**
73
75
  * @deprecated Use aiInput(locatePrompt, opt) instead where opt contains the value
74
76
  */
75
77
  aiInput(value: string, locatePrompt: TUserPrompt, opt?: LocateOption & {
76
78
  autoDismissKeyboard?: boolean;
79
+ } & {
80
+ append?: boolean;
77
81
  }): Promise<any>;
78
82
  aiKeyboardPress(locatePrompt: TUserPrompt, opt: LocateOption & {
79
83
  keyName: string;
@@ -138,8 +138,7 @@ export declare const TUserPromptSchema: z.ZodUnion<[z.ZodString, z.ZodIntersecti
138
138
  }>>]>;
139
139
  export type TMultimodalPrompt = z.infer<typeof TMultimodalPromptSchema>;
140
140
  export type TUserPrompt = z.infer<typeof TUserPromptSchema>;
141
- declare const MidsceneLocationResult: z.ZodObject<{
142
- midscene_location_field_flag: z.ZodLiteral<true>;
141
+ declare const MidsceneLocationInput: z.ZodObject<{
143
142
  prompt: z.ZodUnion<[z.ZodString, z.ZodIntersection<z.ZodObject<{
144
143
  prompt: z.ZodString;
145
144
  }, "strip", z.ZodTypeAny, {
@@ -173,38 +172,8 @@ declare const MidsceneLocationResult: z.ZodObject<{
173
172
  }>>]>;
174
173
  deepThink: z.ZodOptional<z.ZodBoolean>;
175
174
  cacheable: z.ZodOptional<z.ZodBoolean>;
176
- xpath: z.ZodOptional<z.ZodBoolean>;
177
- center: z.ZodTuple<[z.ZodNumber, z.ZodNumber], null>;
178
- rect: z.ZodIntersection<z.ZodIntersection<z.ZodObject<{
179
- left: z.ZodNumber;
180
- top: z.ZodNumber;
181
- }, "strip", z.ZodTypeAny, {
182
- left: number;
183
- top: number;
184
- }, {
185
- left: number;
186
- top: number;
187
- }>, z.ZodObject<{
188
- width: z.ZodNumber;
189
- height: z.ZodNumber;
190
- dpr: z.ZodOptional<z.ZodNumber>;
191
- }, "strip", z.ZodTypeAny, {
192
- width: number;
193
- height: number;
194
- dpr?: number | undefined;
195
- }, {
196
- width: number;
197
- height: number;
198
- dpr?: number | undefined;
199
- }>>, z.ZodObject<{
200
- zoom: z.ZodOptional<z.ZodNumber>;
201
- }, "strip", z.ZodTypeAny, {
202
- zoom?: number | undefined;
203
- }, {
204
- zoom?: number | undefined;
205
- }>>;
175
+ xpath: z.ZodOptional<z.ZodUnion<[z.ZodString, z.ZodBoolean]>>;
206
176
  }, "passthrough", z.ZodTypeAny, z.objectOutputType<{
207
- midscene_location_field_flag: z.ZodLiteral<true>;
208
177
  prompt: z.ZodUnion<[z.ZodString, z.ZodIntersection<z.ZodObject<{
209
178
  prompt: z.ZodString;
210
179
  }, "strip", z.ZodTypeAny, {
@@ -238,37 +207,44 @@ declare const MidsceneLocationResult: z.ZodObject<{
238
207
  }>>]>;
239
208
  deepThink: z.ZodOptional<z.ZodBoolean>;
240
209
  cacheable: z.ZodOptional<z.ZodBoolean>;
241
- xpath: z.ZodOptional<z.ZodBoolean>;
242
- center: z.ZodTuple<[z.ZodNumber, z.ZodNumber], null>;
243
- rect: z.ZodIntersection<z.ZodIntersection<z.ZodObject<{
244
- left: z.ZodNumber;
245
- top: z.ZodNumber;
210
+ xpath: z.ZodOptional<z.ZodUnion<[z.ZodString, z.ZodBoolean]>>;
211
+ }, z.ZodTypeAny, "passthrough">, z.objectInputType<{
212
+ prompt: z.ZodUnion<[z.ZodString, z.ZodIntersection<z.ZodObject<{
213
+ prompt: z.ZodString;
246
214
  }, "strip", z.ZodTypeAny, {
247
- left: number;
248
- top: number;
215
+ prompt: string;
249
216
  }, {
250
- left: number;
251
- top: number;
217
+ prompt: string;
252
218
  }>, z.ZodObject<{
253
- width: z.ZodNumber;
254
- height: z.ZodNumber;
255
- dpr: z.ZodOptional<z.ZodNumber>;
256
- }, "strip", z.ZodTypeAny, {
257
- width: number;
258
- height: number;
259
- dpr?: number | undefined;
260
- }, {
261
- width: number;
262
- height: number;
263
- dpr?: number | undefined;
264
- }>>, z.ZodObject<{
265
- zoom: z.ZodOptional<z.ZodNumber>;
219
+ images: z.ZodOptional<z.ZodOptional<z.ZodArray<z.ZodObject<{
220
+ name: z.ZodString;
221
+ url: z.ZodString;
222
+ }, "strip", z.ZodTypeAny, {
223
+ name: string;
224
+ url: string;
225
+ }, {
226
+ name: string;
227
+ url: string;
228
+ }>, "many">>>;
229
+ convertHttpImage2Base64: z.ZodOptional<z.ZodOptional<z.ZodBoolean>>;
266
230
  }, "strip", z.ZodTypeAny, {
267
- zoom?: number | undefined;
231
+ images?: {
232
+ name: string;
233
+ url: string;
234
+ }[] | undefined;
235
+ convertHttpImage2Base64?: boolean | undefined;
268
236
  }, {
269
- zoom?: number | undefined;
270
- }>>;
271
- }, z.ZodTypeAny, "passthrough">, z.objectInputType<{
237
+ images?: {
238
+ name: string;
239
+ url: string;
240
+ }[] | undefined;
241
+ convertHttpImage2Base64?: boolean | undefined;
242
+ }>>]>;
243
+ deepThink: z.ZodOptional<z.ZodBoolean>;
244
+ cacheable: z.ZodOptional<z.ZodBoolean>;
245
+ xpath: z.ZodOptional<z.ZodUnion<[z.ZodString, z.ZodBoolean]>>;
246
+ }, z.ZodTypeAny, "passthrough">>;
247
+ declare const MidsceneLocationResult: z.ZodObject<{
272
248
  midscene_location_field_flag: z.ZodLiteral<true>;
273
249
  prompt: z.ZodUnion<[z.ZodString, z.ZodIntersection<z.ZodObject<{
274
250
  prompt: z.ZodString;
@@ -333,9 +309,7 @@ declare const MidsceneLocationResult: z.ZodObject<{
333
309
  }, {
334
310
  zoom?: number | undefined;
335
311
  }>>;
336
- }, z.ZodTypeAny, "passthrough">>;
337
- export type MidsceneLocationResultType = z.infer<typeof MidsceneLocationResult>;
338
- export declare const getMidsceneLocationSchema: () => z.ZodObject<{
312
+ }, "passthrough", z.ZodTypeAny, z.objectOutputType<{
339
313
  midscene_location_field_flag: z.ZodLiteral<true>;
340
314
  prompt: z.ZodUnion<[z.ZodString, z.ZodIntersection<z.ZodObject<{
341
315
  prompt: z.ZodString;
@@ -400,7 +374,7 @@ export declare const getMidsceneLocationSchema: () => z.ZodObject<{
400
374
  }, {
401
375
  zoom?: number | undefined;
402
376
  }>>;
403
- }, "passthrough", z.ZodTypeAny, z.objectOutputType<{
377
+ }, z.ZodTypeAny, "passthrough">, z.objectInputType<{
404
378
  midscene_location_field_flag: z.ZodLiteral<true>;
405
379
  prompt: z.ZodUnion<[z.ZodString, z.ZodIntersection<z.ZodObject<{
406
380
  prompt: z.ZodString;
@@ -465,8 +439,14 @@ export declare const getMidsceneLocationSchema: () => z.ZodObject<{
465
439
  }, {
466
440
  zoom?: number | undefined;
467
441
  }>>;
468
- }, z.ZodTypeAny, "passthrough">, z.objectInputType<{
469
- midscene_location_field_flag: z.ZodLiteral<true>;
442
+ }, z.ZodTypeAny, "passthrough">>;
443
+ export type MidsceneLocationResultType = z.infer<typeof MidsceneLocationResult>;
444
+ export type MidsceneLocationInputType = z.infer<typeof MidsceneLocationInput>;
445
+ /**
446
+ * Returns the schema for locator fields.
447
+ * This now returns the input schema which is more permissive and suitable for validation.
448
+ */
449
+ export declare const getMidsceneLocationSchema: () => z.ZodObject<{
470
450
  prompt: z.ZodUnion<[z.ZodString, z.ZodIntersection<z.ZodObject<{
471
451
  prompt: z.ZodString;
472
452
  }, "strip", z.ZodTypeAny, {
@@ -500,40 +480,87 @@ export declare const getMidsceneLocationSchema: () => z.ZodObject<{
500
480
  }>>]>;
501
481
  deepThink: z.ZodOptional<z.ZodBoolean>;
502
482
  cacheable: z.ZodOptional<z.ZodBoolean>;
503
- xpath: z.ZodOptional<z.ZodBoolean>;
504
- center: z.ZodTuple<[z.ZodNumber, z.ZodNumber], null>;
505
- rect: z.ZodIntersection<z.ZodIntersection<z.ZodObject<{
506
- left: z.ZodNumber;
507
- top: z.ZodNumber;
483
+ xpath: z.ZodOptional<z.ZodUnion<[z.ZodString, z.ZodBoolean]>>;
484
+ }, "passthrough", z.ZodTypeAny, z.objectOutputType<{
485
+ prompt: z.ZodUnion<[z.ZodString, z.ZodIntersection<z.ZodObject<{
486
+ prompt: z.ZodString;
508
487
  }, "strip", z.ZodTypeAny, {
509
- left: number;
510
- top: number;
488
+ prompt: string;
511
489
  }, {
512
- left: number;
513
- top: number;
490
+ prompt: string;
514
491
  }>, z.ZodObject<{
515
- width: z.ZodNumber;
516
- height: z.ZodNumber;
517
- dpr: z.ZodOptional<z.ZodNumber>;
492
+ images: z.ZodOptional<z.ZodOptional<z.ZodArray<z.ZodObject<{
493
+ name: z.ZodString;
494
+ url: z.ZodString;
495
+ }, "strip", z.ZodTypeAny, {
496
+ name: string;
497
+ url: string;
498
+ }, {
499
+ name: string;
500
+ url: string;
501
+ }>, "many">>>;
502
+ convertHttpImage2Base64: z.ZodOptional<z.ZodOptional<z.ZodBoolean>>;
518
503
  }, "strip", z.ZodTypeAny, {
519
- width: number;
520
- height: number;
521
- dpr?: number | undefined;
504
+ images?: {
505
+ name: string;
506
+ url: string;
507
+ }[] | undefined;
508
+ convertHttpImage2Base64?: boolean | undefined;
522
509
  }, {
523
- width: number;
524
- height: number;
525
- dpr?: number | undefined;
526
- }>>, z.ZodObject<{
527
- zoom: z.ZodOptional<z.ZodNumber>;
510
+ images?: {
511
+ name: string;
512
+ url: string;
513
+ }[] | undefined;
514
+ convertHttpImage2Base64?: boolean | undefined;
515
+ }>>]>;
516
+ deepThink: z.ZodOptional<z.ZodBoolean>;
517
+ cacheable: z.ZodOptional<z.ZodBoolean>;
518
+ xpath: z.ZodOptional<z.ZodUnion<[z.ZodString, z.ZodBoolean]>>;
519
+ }, z.ZodTypeAny, "passthrough">, z.objectInputType<{
520
+ prompt: z.ZodUnion<[z.ZodString, z.ZodIntersection<z.ZodObject<{
521
+ prompt: z.ZodString;
528
522
  }, "strip", z.ZodTypeAny, {
529
- zoom?: number | undefined;
523
+ prompt: string;
530
524
  }, {
531
- zoom?: number | undefined;
532
- }>>;
525
+ prompt: string;
526
+ }>, z.ZodObject<{
527
+ images: z.ZodOptional<z.ZodOptional<z.ZodArray<z.ZodObject<{
528
+ name: z.ZodString;
529
+ url: z.ZodString;
530
+ }, "strip", z.ZodTypeAny, {
531
+ name: string;
532
+ url: string;
533
+ }, {
534
+ name: string;
535
+ url: string;
536
+ }>, "many">>>;
537
+ convertHttpImage2Base64: z.ZodOptional<z.ZodOptional<z.ZodBoolean>>;
538
+ }, "strip", z.ZodTypeAny, {
539
+ images?: {
540
+ name: string;
541
+ url: string;
542
+ }[] | undefined;
543
+ convertHttpImage2Base64?: boolean | undefined;
544
+ }, {
545
+ images?: {
546
+ name: string;
547
+ url: string;
548
+ }[] | undefined;
549
+ convertHttpImage2Base64?: boolean | undefined;
550
+ }>>]>;
551
+ deepThink: z.ZodOptional<z.ZodBoolean>;
552
+ cacheable: z.ZodOptional<z.ZodBoolean>;
553
+ xpath: z.ZodOptional<z.ZodUnion<[z.ZodString, z.ZodBoolean]>>;
533
554
  }, z.ZodTypeAny, "passthrough">>;
534
555
  export declare const ifMidsceneLocatorField: (field: any) => boolean;
535
556
  export declare const dumpMidsceneLocatorField: (field: any) => string;
536
557
  export declare const findAllMidsceneLocatorField: (zodType?: z.ZodType<any>, requiredOnly?: boolean) => string[];
537
558
  export declare const dumpActionParam: (jsonObject: Record<string, any>, zodSchema: z.ZodType<any>) => Record<string, any>;
538
559
  export declare const loadActionParam: (jsonObject: Record<string, any>, zodSchema: z.ZodType<any>) => Record<string, any>;
560
+ /**
561
+ * Parse and validate action parameters using Zod schema.
562
+ * All fields are validated through Zod, including locator fields which have their own schema.
563
+ * Default values defined in the schema are automatically applied.
564
+ */
565
+ export declare const parseActionParam: (rawParam: Record<string, any>, zodSchema: z.ZodType<any>) => Record<string, any>;
539
566
  export {};
@@ -10,4 +10,4 @@ export { adaptBboxToRect } from './common';
10
10
  export { uiTarsPlanning, resizeImageForUiTars } from './ui-tars-planning';
11
11
  export { ConversationHistory, type ConversationHistoryOptions, } from './conversation-history';
12
12
  export { AIActionType, type AIArgs } from './common';
13
- export { getMidsceneLocationSchema, type MidsceneLocationResultType, PointSchema, SizeSchema, RectSchema, TMultimodalPromptSchema, TUserPromptSchema, type TMultimodalPrompt, type TUserPrompt, findAllMidsceneLocatorField, dumpActionParam, loadActionParam, } from './common';
13
+ export { getMidsceneLocationSchema, type MidsceneLocationResultType, PointSchema, SizeSchema, RectSchema, TMultimodalPromptSchema, TUserPromptSchema, type TMultimodalPrompt, type TUserPrompt, findAllMidsceneLocatorField, dumpActionParam, loadActionParam, parseActionParam, } from './common';