@midscene/core 0.24.2-beta-20250731102409.0 → 0.24.2-beta-20250731132300.0

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.
Files changed (32) hide show
  1. package/dist/es/ai-model.d.ts +3 -3
  2. package/dist/es/ai-model.js +1 -1
  3. package/dist/es/{chunk-YCHAOUOW.js → chunk-FKQMUAXP.js} +46 -15
  4. package/dist/es/chunk-FKQMUAXP.js.map +1 -0
  5. package/dist/es/{chunk-TYKJHDNK.js → chunk-IMZJSEAX.js} +3 -3
  6. package/dist/es/index.d.ts +4 -4
  7. package/dist/es/index.js +2 -2
  8. package/dist/es/{llm-planning-4bc82162.d.ts → llm-planning-d7096b0d.d.ts} +1 -1
  9. package/dist/es/{types-93fa8419.d.ts → types-d836fa73.d.ts} +16 -2
  10. package/dist/es/utils.d.ts +1 -1
  11. package/dist/es/utils.js +1 -1
  12. package/dist/lib/ai-model.d.ts +3 -3
  13. package/dist/lib/ai-model.js +2 -2
  14. package/dist/lib/{chunk-YCHAOUOW.js → chunk-FKQMUAXP.js} +55 -24
  15. package/dist/lib/chunk-FKQMUAXP.js.map +1 -0
  16. package/dist/lib/{chunk-TYKJHDNK.js → chunk-IMZJSEAX.js} +3 -3
  17. package/dist/lib/index.d.ts +4 -4
  18. package/dist/lib/index.js +12 -12
  19. package/dist/lib/{llm-planning-4bc82162.d.ts → llm-planning-d7096b0d.d.ts} +1 -1
  20. package/dist/{types/types-93fa8419.d.ts → lib/types-d836fa73.d.ts} +16 -2
  21. package/dist/lib/utils.d.ts +1 -1
  22. package/dist/lib/utils.js +2 -2
  23. package/dist/types/ai-model.d.ts +3 -3
  24. package/dist/types/index.d.ts +4 -4
  25. package/dist/types/{llm-planning-4bc82162.d.ts → llm-planning-d7096b0d.d.ts} +1 -1
  26. package/dist/{lib/types-93fa8419.d.ts → types/types-d836fa73.d.ts} +16 -2
  27. package/dist/types/utils.d.ts +1 -1
  28. package/package.json +3 -3
  29. package/dist/es/chunk-YCHAOUOW.js.map +0 -1
  30. package/dist/lib/chunk-YCHAOUOW.js.map +0 -1
  31. /package/dist/es/{chunk-TYKJHDNK.js.map → chunk-IMZJSEAX.js.map} +0 -0
  32. /package/dist/lib/{chunk-TYKJHDNK.js.map → chunk-IMZJSEAX.js.map} +0 -0
@@ -1,7 +1,7 @@
1
- import { E as ExecutionTask, a as ExecutionTaskProgressOptions, b as ExecutionTaskApply, c as ExecutionDump, U as UIContext, I as InsightAction, D as DumpSubscriber, d as InsightTaskInfo, e as InsightOptions, f as DetailedLocateParam, L as LocateResult, g as InsightExtractOption, h as InsightAssertionResponse, A as AIDescribeElementResponse } from './types-93fa8419.js';
2
- export { w as AIAssertionResponse, u as AIDataExtractionResponse, s as AIElementCoordinatesResponse, r as AIElementLocatorResponse, t as AIElementResponse, n as AIResponseFormat, v as AISectionLocatorResponse, q as AISingleElementResponse, o as AISingleElementResponseById, p as AISingleElementResponseByPosition, m as AIUsageInfo, S as AgentAssertOpt, z as AgentDescribeElementAtPointResult, Q as AgentWaitForOpt, a5 as BaseAgentParserOpt, C as CallAIFn, az as CodeGenerationChunk, a4 as Color, H as DumpMeta, N as ElementById, B as EnsureObject, a8 as ExecutionRecorderItem, aq as ExecutionTaskAction, ap as ExecutionTaskActionApply, ab as ExecutionTaskHitBy, ao as ExecutionTaskInsightAssertion, an as ExecutionTaskInsightAssertionApply, am as ExecutionTaskInsightAssertionParam, af as ExecutionTaskInsightDumpLog, ah as ExecutionTaskInsightLocate, ag as ExecutionTaskInsightLocateApply, ae as ExecutionTaskInsightLocateOutput, ad as ExecutionTaskInsightLocateParam, al as ExecutionTaskInsightQuery, ak as ExecutionTaskInsightQueryApply, aj as ExecutionTaskInsightQueryOutput, ai as ExecutionTaskInsightQueryParam, as as ExecutionTaskLog, ar as ExecutionTaskLogApply, au as ExecutionTaskPlanning, at as ExecutionTaskPlanningApply, ac as ExecutionTaskReturn, a9 as ExecutionTaskType, aa as ExecutorContext, aZ as FreeFn, av as GroupedActionDump, J as InsightDump, F as InsightExtractParam, K as LiteUISection, aB as LocateOption, G as LocateResultElement, y as LocateValidatorResult, x as LocatorValidatorOption, b0 as MidsceneYamlConfig, b1 as MidsceneYamlConfigOutput, l as MidsceneYamlConfigResult, j as MidsceneYamlFlowItem, aI as MidsceneYamlFlowItemAIAction, aN as MidsceneYamlFlowItemAIAsk, aJ as MidsceneYamlFlowItemAIAssert, aO as MidsceneYamlFlowItemAIBoolean, aS as MidsceneYamlFlowItemAIHover, aT as MidsceneYamlFlowItemAIInput, aU as MidsceneYamlFlowItemAIKeyboardPress, aP as MidsceneYamlFlowItemAILocate, aL as MidsceneYamlFlowItemAINumber, aK as MidsceneYamlFlowItemAIQuery, k as MidsceneYamlFlowItemAIRightClick, aV as MidsceneYamlFlowItemAIScroll, aM as MidsceneYamlFlowItemAIString, aR as MidsceneYamlFlowItemAITap, aQ as MidsceneYamlFlowItemAIWaitFor, aW as MidsceneYamlFlowItemEvaluateJavaScript, aY as MidsceneYamlFlowItemLogScreenshot, aX as MidsceneYamlFlowItemSleep, M as MidsceneYamlScript, aG as MidsceneYamlScriptAndroidEnv, aH as MidsceneYamlScriptEnv, aE as MidsceneYamlScriptEnvBase, aF as MidsceneYamlScriptWebEnv, i as MidsceneYamlTask, O as OnTaskStartTip, aw as PageType, P as PartialInsightDumpFromSDK, W as PlanningAIResponse, V as PlanningAction, a0 as PlanningActionParamAssert, a2 as PlanningActionParamError, Y as PlanningActionParamHover, _ as PlanningActionParamInputOrKeyPress, Z as PlanningActionParamRightClick, $ as PlanningActionParamScroll, a1 as PlanningActionParamSleep, X as PlanningActionParamTap, a3 as PlanningActionParamWaitFor, T as PlanningLocateParam, a7 as PlaywrightParserOpt, a6 as PuppeteerParserOpt, aC as ReferenceImage, R as ReportDumpWithAttributes, a$ as ScriptPlayerStatusValue, a_ as ScriptPlayerTaskStatus, aA as StreamingAIResponse, ay as StreamingCallback, ax as StreamingCodeGenerationOptions, aD as scrollParam } from './types-93fa8419.js';
3
- import { c as callAiFn } from './llm-planning-4bc82162.js';
4
- export { a as AiAssert, A as AiLocateElement, d as describeUserPage, p as plan } from './llm-planning-4bc82162.js';
1
+ import { E as ExecutionTask, a as ExecutionTaskProgressOptions, b as ExecutionTaskApply, c as ExecutionDump, U as UIContext, I as InsightAction, D as DumpSubscriber, d as InsightTaskInfo, e as InsightOptions, f as DetailedLocateParam, L as LocateResult, g as InsightExtractOption, h as InsightAssertionResponse, A as AIDescribeElementResponse } from './types-d836fa73.js';
2
+ export { w as AIAssertionResponse, u as AIDataExtractionResponse, s as AIElementCoordinatesResponse, r as AIElementLocatorResponse, t as AIElementResponse, n as AIResponseFormat, v as AISectionLocatorResponse, q as AISingleElementResponse, o as AISingleElementResponseById, p as AISingleElementResponseByPosition, m as AIUsageInfo, S as AgentAssertOpt, z as AgentDescribeElementAtPointResult, Q as AgentWaitForOpt, a7 as BaseAgentParserOpt, C as CallAIFn, aB as CodeGenerationChunk, a6 as Color, H as DumpMeta, N as ElementById, B as EnsureObject, aa as ExecutionRecorderItem, as as ExecutionTaskAction, ar as ExecutionTaskActionApply, ad as ExecutionTaskHitBy, aq as ExecutionTaskInsightAssertion, ap as ExecutionTaskInsightAssertionApply, ao as ExecutionTaskInsightAssertionParam, ah as ExecutionTaskInsightDumpLog, aj as ExecutionTaskInsightLocate, ai as ExecutionTaskInsightLocateApply, ag as ExecutionTaskInsightLocateOutput, af as ExecutionTaskInsightLocateParam, an as ExecutionTaskInsightQuery, am as ExecutionTaskInsightQueryApply, al as ExecutionTaskInsightQueryOutput, ak as ExecutionTaskInsightQueryParam, au as ExecutionTaskLog, at as ExecutionTaskLogApply, aw as ExecutionTaskPlanning, av as ExecutionTaskPlanningApply, ae as ExecutionTaskReturn, ab as ExecutionTaskType, ac as ExecutorContext, a$ as FreeFn, ax as GroupedActionDump, J as InsightDump, F as InsightExtractParam, K as LiteUISection, aD as LocateOption, G as LocateResultElement, y as LocateValidatorResult, x as LocatorValidatorOption, b2 as MidsceneYamlConfig, b3 as MidsceneYamlConfigOutput, l as MidsceneYamlConfigResult, j as MidsceneYamlFlowItem, aK as MidsceneYamlFlowItemAIAction, aP as MidsceneYamlFlowItemAIAsk, aL as MidsceneYamlFlowItemAIAssert, aQ as MidsceneYamlFlowItemAIBoolean, aU as MidsceneYamlFlowItemAIHover, aV as MidsceneYamlFlowItemAIInput, aW as MidsceneYamlFlowItemAIKeyboardPress, aR as MidsceneYamlFlowItemAILocate, aN as MidsceneYamlFlowItemAINumber, aM as MidsceneYamlFlowItemAIQuery, k as MidsceneYamlFlowItemAIRightClick, aX as MidsceneYamlFlowItemAIScroll, aO as MidsceneYamlFlowItemAIString, aT as MidsceneYamlFlowItemAITap, aS as MidsceneYamlFlowItemAIWaitFor, aY as MidsceneYamlFlowItemEvaluateJavaScript, a_ as MidsceneYamlFlowItemLogScreenshot, aZ as MidsceneYamlFlowItemSleep, M as MidsceneYamlScript, aI as MidsceneYamlScriptAndroidEnv, aJ as MidsceneYamlScriptEnv, aG as MidsceneYamlScriptEnvBase, aH as MidsceneYamlScriptWebEnv, i as MidsceneYamlTask, O as OnTaskStartTip, ay as PageType, P as PartialInsightDumpFromSDK, W as PlanningAIResponse, V as PlanningAction, a4 as PlanningActionParamAndroidLongPress, a5 as PlanningActionParamAndroidPull, a0 as PlanningActionParamAssert, a2 as PlanningActionParamError, Y as PlanningActionParamHover, _ as PlanningActionParamInputOrKeyPress, Z as PlanningActionParamRightClick, $ as PlanningActionParamScroll, a1 as PlanningActionParamSleep, X as PlanningActionParamTap, a3 as PlanningActionParamWaitFor, T as PlanningLocateParam, a9 as PlaywrightParserOpt, a8 as PuppeteerParserOpt, aE as ReferenceImage, R as ReportDumpWithAttributes, b1 as ScriptPlayerStatusValue, b0 as ScriptPlayerTaskStatus, aC as StreamingAIResponse, aA as StreamingCallback, az as StreamingCodeGenerationOptions, aF as scrollParam } from './types-d836fa73.js';
3
+ import { c as callAiFn } from './llm-planning-d7096b0d.js';
4
+ export { a as AiAssert, A as AiLocateElement, d as describeUserPage, p as plan } from './llm-planning-d7096b0d.js';
5
5
  import { BaseElement, Rect } from '@midscene/shared/types';
6
6
  export { BaseElement, ElementTreeNode, Point, Rect, Size } from '@midscene/shared/types';
7
7
  export { getVersion } from './utils.js';
package/dist/lib/index.js CHANGED
@@ -1,6 +1,6 @@
1
1
  "use strict";Object.defineProperty(exports, "__esModule", {value: true}); function _optionalChain(ops) { let lastAccessLHS = undefined; let value = ops[0]; let i = 1; while (i < ops.length) { const op = ops[i]; const fn = ops[i + 1]; i += 2; if ((op === 'optionalAccess' || op === 'optionalCall') && value == null) { return undefined; } if (op === 'access' || op === 'optionalAccess') { lastAccessLHS = value; value = fn(value); } else if (op === 'call' || op === 'optionalCall') { value = fn((...args) => value.call(lastAccessLHS, ...args)); lastAccessLHS = undefined; } } return value; }
2
2
 
3
- var _chunkTYKJHDNKjs = require('./chunk-TYKJHDNK.js');
3
+ var _chunkIMZJSEAXjs = require('./chunk-IMZJSEAX.js');
4
4
 
5
5
 
6
6
 
@@ -11,7 +11,7 @@ var _chunkTYKJHDNKjs = require('./chunk-TYKJHDNK.js');
11
11
 
12
12
 
13
13
 
14
- var _chunkYCHAOUOWjs = require('./chunk-YCHAOUOW.js');
14
+ var _chunkFKQMUAXPjs = require('./chunk-FKQMUAXP.js');
15
15
 
16
16
  // src/ai-model/action-executor.ts
17
17
 
@@ -172,7 +172,7 @@ ${_optionalChain([this, 'access', _7 => _7.latestErrorTask, 'call', _8 => _8(),
172
172
  }
173
173
  }
174
174
  const dumpData = {
175
- sdkVersion: _chunkTYKJHDNKjs.getVersion.call(void 0, ),
175
+ sdkVersion: _chunkIMZJSEAXjs.getVersion.call(void 0, ),
176
176
  model_name: _env.getAIConfig.call(void 0, _env.MIDSCENE_MODEL_NAME) || "",
177
177
  model_description: modelDescription,
178
178
  logTime: Date.now(),
@@ -220,7 +220,7 @@ var _logger = require('@midscene/shared/logger');
220
220
 
221
221
  function emitInsightDump(data, dumpSubscriber) {
222
222
  const baseData = {
223
- sdkVersion: _chunkTYKJHDNKjs.getVersion.call(void 0, ),
223
+ sdkVersion: _chunkIMZJSEAXjs.getVersion.call(void 0, ),
224
224
  logTime: Date.now(),
225
225
  model_name: _env.getAIConfig.call(void 0, _env.MIDSCENE_MODEL_NAME) || ""
226
226
  };
@@ -236,7 +236,7 @@ function emitInsightDump(data, dumpSubscriber) {
236
236
  var debug = _logger.getDebug.call(void 0, "ai:insight");
237
237
  var Insight = class {
238
238
  constructor(context, opt) {
239
- this.aiVendorFn = _chunkYCHAOUOWjs.callAiFn;
239
+ this.aiVendorFn = _chunkFKQMUAXPjs.callAiFn;
240
240
  _utils.assert.call(void 0, context, "context is required for Insight");
241
241
  if (typeof context === "function") {
242
242
  this.contextRetrieverFn = context;
@@ -279,7 +279,7 @@ var Insight = class {
279
279
  let searchAreaUsage = void 0;
280
280
  let searchAreaResponse = void 0;
281
281
  if (searchAreaPrompt) {
282
- searchAreaResponse = await _chunkYCHAOUOWjs.AiLocateSection.call(void 0, {
282
+ searchAreaResponse = await _chunkFKQMUAXPjs.AiLocateSection.call(void 0, {
283
283
  context,
284
284
  sectionDescription: searchAreaPrompt
285
285
  });
@@ -299,7 +299,7 @@ var Insight = class {
299
299
  rawResponse,
300
300
  usage,
301
301
  isOrderSensitive
302
- } = await _chunkYCHAOUOWjs.AiLocateElement.call(void 0, {
302
+ } = await _chunkFKQMUAXPjs.AiLocateElement.call(void 0, {
303
303
  callAI: callAI || this.aiVendorFn,
304
304
  context,
305
305
  targetElementDescription: queryPrompt,
@@ -388,7 +388,7 @@ ${parseResult.errors.join("\n")}`;
388
388
  this.onceDumpUpdatedFn = void 0;
389
389
  const context = await this.contextRetrieverFn("extract");
390
390
  const startTime = Date.now();
391
- const { parseResult, usage } = await _chunkYCHAOUOWjs.AiExtractElementInfo.call(void 0, {
391
+ const { parseResult, usage } = await _chunkFKQMUAXPjs.AiExtractElementInfo.call(void 0, {
392
392
  context,
393
393
  dataQuery: dataDemand,
394
394
  extractOption: opt
@@ -440,7 +440,7 @@ ${parseResult.errors.join("\n")}`;
440
440
  this.onceDumpUpdatedFn = void 0;
441
441
  const context = await this.contextRetrieverFn("assert");
442
442
  const startTime = Date.now();
443
- const assertResult = await _chunkYCHAOUOWjs.AiAssert.call(void 0, {
443
+ const assertResult = await _chunkFKQMUAXPjs.AiAssert.call(void 0, {
444
444
  assertion,
445
445
  context
446
446
  });
@@ -494,7 +494,7 @@ ${parseResult.errors.join("\n")}`;
494
494
  borderThickness: 3
495
495
  });
496
496
  if (_optionalChain([opt, 'optionalAccess', _22 => _22.deepThink])) {
497
- const searchArea = _chunkYCHAOUOWjs.expandSearchArea.call(void 0, targetRect, context.size);
497
+ const searchArea = _chunkFKQMUAXPjs.expandSearchArea.call(void 0, targetRect, context.size);
498
498
  debug("describe: set searchArea", searchArea);
499
499
  imagePayload = await _img.cropByRect.call(void 0,
500
500
  imagePayload,
@@ -517,7 +517,7 @@ ${parseResult.errors.join("\n")}`;
517
517
  ]
518
518
  }
519
519
  ];
520
- const callAIFn = this.aiVendorFn || _chunkYCHAOUOWjs.callToGetJSONObject;
520
+ const callAIFn = this.aiVendorFn || _chunkFKQMUAXPjs.callToGetJSONObject;
521
521
  const res = await callAIFn(msgs, 4 /* DESCRIBE_ELEMENT */);
522
522
  const { content } = res;
523
523
  _utils.assert.call(void 0, !content.error, `describe failed: ${content.error}`);
@@ -540,6 +540,6 @@ var src_default = Insight;
540
540
 
541
541
 
542
542
 
543
- exports.AiAssert = _chunkYCHAOUOWjs.AiAssert; exports.AiLocateElement = _chunkYCHAOUOWjs.AiLocateElement; exports.Executor = Executor; exports.Insight = Insight; exports.MIDSCENE_MODEL_NAME = _env.MIDSCENE_MODEL_NAME; exports.default = src_default; exports.describeUserPage = _chunkYCHAOUOWjs.describeUserPage; exports.getAIConfig = _env.getAIConfig; exports.getVersion = _chunkTYKJHDNKjs.getVersion; exports.plan = _chunkYCHAOUOWjs.plan;
543
+ exports.AiAssert = _chunkFKQMUAXPjs.AiAssert; exports.AiLocateElement = _chunkFKQMUAXPjs.AiLocateElement; exports.Executor = Executor; exports.Insight = Insight; exports.MIDSCENE_MODEL_NAME = _env.MIDSCENE_MODEL_NAME; exports.default = src_default; exports.describeUserPage = _chunkFKQMUAXPjs.describeUserPage; exports.getAIConfig = _env.getAIConfig; exports.getVersion = _chunkIMZJSEAXjs.getVersion; exports.plan = _chunkFKQMUAXPjs.plan;
544
544
 
545
545
  //# sourceMappingURL=index.js.map
@@ -1,4 +1,4 @@
1
- import { m as AIUsageInfo, U as UIContext, aC as ReferenceImage, r as AIElementLocatorResponse, N as ElementById, g as InsightExtractOption, u as AIDataExtractionResponse, w as AIAssertionResponse, aw as PageType, W as PlanningAIResponse } from './types-93fa8419.js';
1
+ import { m as AIUsageInfo, U as UIContext, aE as ReferenceImage, r as AIElementLocatorResponse, N as ElementById, g as InsightExtractOption, u as AIDataExtractionResponse, w as AIAssertionResponse, ay as PageType, W as PlanningAIResponse } from './types-d836fa73.js';
2
2
  import { Rect, ElementTreeNode, BaseElement } from '@midscene/shared/types';
3
3
  import { ChatCompletionSystemMessageParam, ChatCompletionUserMessageParam } from 'openai/resources';
4
4
 
@@ -357,7 +357,7 @@ interface PlanningLocateParam extends DetailedLocateParam {
357
357
  }
358
358
  interface PlanningAction<ParamType = any> {
359
359
  thought?: string;
360
- type: 'Locate' | 'Tap' | 'RightClick' | 'Hover' | 'Drag' | 'Input' | 'KeyboardPress' | 'Scroll' | 'Error' | 'ExpectedFalsyCondition' | 'Assert' | 'AssertWithoutThrow' | 'Sleep' | 'Finished' | 'AndroidBackButton' | 'AndroidHomeButton' | 'AndroidRecentAppsButton';
360
+ type: 'Locate' | 'Tap' | 'RightClick' | 'Hover' | 'Drag' | 'Input' | 'KeyboardPress' | 'Scroll' | 'Error' | 'ExpectedFalsyCondition' | 'Assert' | 'AssertWithoutThrow' | 'Sleep' | 'Finished' | 'AndroidBackButton' | 'AndroidHomeButton' | 'AndroidRecentAppsButton' | 'AndroidLongPress' | 'AndroidPull';
361
361
  param: ParamType;
362
362
  locate?: PlanningLocateParam | null;
363
363
  }
@@ -393,6 +393,20 @@ interface PlanningActionParamError {
393
393
  type PlanningActionParamWaitFor = AgentWaitForOpt & {
394
394
  assertion: string;
395
395
  };
396
+ interface PlanningActionParamAndroidLongPress {
397
+ x: number;
398
+ y: number;
399
+ duration?: number;
400
+ }
401
+ interface PlanningActionParamAndroidPull {
402
+ direction: 'up' | 'down';
403
+ startPoint?: {
404
+ x: number;
405
+ y: number;
406
+ };
407
+ distance?: number;
408
+ duration?: number;
409
+ }
396
410
  /**
397
411
  * misc
398
412
  */
@@ -529,4 +543,4 @@ interface StreamingAIResponse {
529
543
  isStreamed: boolean;
530
544
  }
531
545
 
532
- export { type PlanningActionParamScroll as $, type AIDescribeElementResponse as A, type EnsureObject as B, type CallAIFn as C, type DumpSubscriber as D, type ExecutionTask as E, type InsightExtractParam as F, type LocateResultElement as G, type DumpMeta as H, type InsightAction as I, type InsightDump as J, type LiteUISection as K, type LocateResult as L, type MidsceneYamlScript as M, type ElementById as N, type OnTaskStartTip as O, type PartialInsightDumpFromSDK as P, type AgentWaitForOpt as Q, type ReportDumpWithAttributes as R, type AgentAssertOpt as S, type PlanningLocateParam as T, UIContext as U, type PlanningAction as V, type PlanningAIResponse as W, type PlanningActionParamTap as X, type PlanningActionParamHover as Y, type PlanningActionParamRightClick as Z, type PlanningActionParamInputOrKeyPress as _, type ExecutionTaskProgressOptions as a, type ScriptPlayerStatusValue as a$, type PlanningActionParamAssert as a0, type PlanningActionParamSleep as a1, type PlanningActionParamError as a2, type PlanningActionParamWaitFor as a3, type Color as a4, type BaseAgentParserOpt as a5, type PuppeteerParserOpt as a6, type PlaywrightParserOpt as a7, type ExecutionRecorderItem as a8, type ExecutionTaskType as a9, type StreamingAIResponse as aA, type LocateOption as aB, type ReferenceImage as aC, type scrollParam as aD, type MidsceneYamlScriptEnvBase as aE, type MidsceneYamlScriptWebEnv as aF, type MidsceneYamlScriptAndroidEnv as aG, type MidsceneYamlScriptEnv as aH, type MidsceneYamlFlowItemAIAction as aI, type MidsceneYamlFlowItemAIAssert as aJ, type MidsceneYamlFlowItemAIQuery as aK, type MidsceneYamlFlowItemAINumber as aL, type MidsceneYamlFlowItemAIString as aM, type MidsceneYamlFlowItemAIAsk as aN, type MidsceneYamlFlowItemAIBoolean as aO, type MidsceneYamlFlowItemAILocate as aP, type MidsceneYamlFlowItemAIWaitFor as aQ, type MidsceneYamlFlowItemAITap as aR, type MidsceneYamlFlowItemAIHover as aS, type MidsceneYamlFlowItemAIInput as aT, type MidsceneYamlFlowItemAIKeyboardPress as aU, type MidsceneYamlFlowItemAIScroll as aV, type MidsceneYamlFlowItemEvaluateJavaScript as aW, type MidsceneYamlFlowItemSleep as aX, type MidsceneYamlFlowItemLogScreenshot as aY, type FreeFn as aZ, type ScriptPlayerTaskStatus as a_, type ExecutorContext as aa, type ExecutionTaskHitBy as ab, type ExecutionTaskReturn as ac, type ExecutionTaskInsightLocateParam as ad, type ExecutionTaskInsightLocateOutput as ae, type ExecutionTaskInsightDumpLog as af, type ExecutionTaskInsightLocateApply as ag, type ExecutionTaskInsightLocate as ah, type ExecutionTaskInsightQueryParam as ai, type ExecutionTaskInsightQueryOutput as aj, type ExecutionTaskInsightQueryApply as ak, type ExecutionTaskInsightQuery as al, type ExecutionTaskInsightAssertionParam as am, type ExecutionTaskInsightAssertionApply as an, type ExecutionTaskInsightAssertion as ao, type ExecutionTaskActionApply as ap, type ExecutionTaskAction as aq, type ExecutionTaskLogApply as ar, type ExecutionTaskLog as as, type ExecutionTaskPlanningApply as at, type ExecutionTaskPlanning as au, type GroupedActionDump as av, type PageType as aw, type StreamingCodeGenerationOptions as ax, type StreamingCallback as ay, type CodeGenerationChunk as az, type ExecutionTaskApply as b, type MidsceneYamlConfig as b0, type MidsceneYamlConfigOutput as b1, type ExecutionDump as c, type InsightTaskInfo as d, type InsightOptions as e, type DetailedLocateParam as f, type InsightExtractOption as g, type InsightAssertionResponse as h, type MidsceneYamlTask as i, type MidsceneYamlFlowItem as j, type MidsceneYamlFlowItemAIRightClick as k, type MidsceneYamlConfigResult as l, type AIUsageInfo as m, AIResponseFormat as n, type AISingleElementResponseById as o, type AISingleElementResponseByPosition as p, type AISingleElementResponse as q, type AIElementLocatorResponse as r, type AIElementCoordinatesResponse as s, type AIElementResponse as t, type AIDataExtractionResponse as u, type AISectionLocatorResponse as v, type AIAssertionResponse as w, type LocatorValidatorOption as x, type LocateValidatorResult as y, type AgentDescribeElementAtPointResult as z };
546
+ export { type PlanningActionParamScroll as $, type AIDescribeElementResponse as A, type EnsureObject as B, type CallAIFn as C, type DumpSubscriber as D, type ExecutionTask as E, type InsightExtractParam as F, type LocateResultElement as G, type DumpMeta as H, type InsightAction as I, type InsightDump as J, type LiteUISection as K, type LocateResult as L, type MidsceneYamlScript as M, type ElementById as N, type OnTaskStartTip as O, type PartialInsightDumpFromSDK as P, type AgentWaitForOpt as Q, type ReportDumpWithAttributes as R, type AgentAssertOpt as S, type PlanningLocateParam as T, UIContext as U, type PlanningAction as V, type PlanningAIResponse as W, type PlanningActionParamTap as X, type PlanningActionParamHover as Y, type PlanningActionParamRightClick as Z, type PlanningActionParamInputOrKeyPress as _, type ExecutionTaskProgressOptions as a, type FreeFn as a$, type PlanningActionParamAssert as a0, type PlanningActionParamSleep as a1, type PlanningActionParamError as a2, type PlanningActionParamWaitFor as a3, type PlanningActionParamAndroidLongPress as a4, type PlanningActionParamAndroidPull as a5, type Color as a6, type BaseAgentParserOpt as a7, type PuppeteerParserOpt as a8, type PlaywrightParserOpt as a9, type StreamingCallback as aA, type CodeGenerationChunk as aB, type StreamingAIResponse as aC, type LocateOption as aD, type ReferenceImage as aE, type scrollParam as aF, type MidsceneYamlScriptEnvBase as aG, type MidsceneYamlScriptWebEnv as aH, type MidsceneYamlScriptAndroidEnv as aI, type MidsceneYamlScriptEnv as aJ, type MidsceneYamlFlowItemAIAction as aK, type MidsceneYamlFlowItemAIAssert as aL, type MidsceneYamlFlowItemAIQuery as aM, type MidsceneYamlFlowItemAINumber as aN, type MidsceneYamlFlowItemAIString as aO, type MidsceneYamlFlowItemAIAsk as aP, type MidsceneYamlFlowItemAIBoolean as aQ, type MidsceneYamlFlowItemAILocate as aR, type MidsceneYamlFlowItemAIWaitFor as aS, type MidsceneYamlFlowItemAITap as aT, type MidsceneYamlFlowItemAIHover as aU, type MidsceneYamlFlowItemAIInput as aV, type MidsceneYamlFlowItemAIKeyboardPress as aW, type MidsceneYamlFlowItemAIScroll as aX, type MidsceneYamlFlowItemEvaluateJavaScript as aY, type MidsceneYamlFlowItemSleep as aZ, type MidsceneYamlFlowItemLogScreenshot as a_, type ExecutionRecorderItem as aa, type ExecutionTaskType as ab, type ExecutorContext as ac, type ExecutionTaskHitBy as ad, type ExecutionTaskReturn as ae, type ExecutionTaskInsightLocateParam as af, type ExecutionTaskInsightLocateOutput as ag, type ExecutionTaskInsightDumpLog as ah, type ExecutionTaskInsightLocateApply as ai, type ExecutionTaskInsightLocate as aj, type ExecutionTaskInsightQueryParam as ak, type ExecutionTaskInsightQueryOutput as al, type ExecutionTaskInsightQueryApply as am, type ExecutionTaskInsightQuery as an, type ExecutionTaskInsightAssertionParam as ao, type ExecutionTaskInsightAssertionApply as ap, type ExecutionTaskInsightAssertion as aq, type ExecutionTaskActionApply as ar, type ExecutionTaskAction as as, type ExecutionTaskLogApply as at, type ExecutionTaskLog as au, type ExecutionTaskPlanningApply as av, type ExecutionTaskPlanning as aw, type GroupedActionDump as ax, type PageType as ay, type StreamingCodeGenerationOptions as az, type ExecutionTaskApply as b, type ScriptPlayerTaskStatus as b0, type ScriptPlayerStatusValue as b1, type MidsceneYamlConfig as b2, type MidsceneYamlConfigOutput as b3, type ExecutionDump as c, type InsightTaskInfo as d, type InsightOptions as e, type DetailedLocateParam as f, type InsightExtractOption as g, type InsightAssertionResponse as h, type MidsceneYamlTask as i, type MidsceneYamlFlowItem as j, type MidsceneYamlFlowItemAIRightClick as k, type MidsceneYamlConfigResult as l, type AIUsageInfo as m, AIResponseFormat as n, type AISingleElementResponseById as o, type AISingleElementResponseByPosition as p, type AISingleElementResponse as q, type AIElementLocatorResponse as r, type AIElementCoordinatesResponse as s, type AIElementResponse as t, type AIDataExtractionResponse as u, type AISectionLocatorResponse as v, type AIAssertionResponse as w, type LocatorValidatorOption as x, type LocateValidatorResult as y, type AgentDescribeElementAtPointResult as z };
@@ -1,4 +1,4 @@
1
- import { R as ReportDumpWithAttributes } from './types-93fa8419.js';
1
+ import { R as ReportDumpWithAttributes } from './types-d836fa73.js';
2
2
  import { Rect } from '@midscene/shared/types';
3
3
  import '@midscene/shared/constants';
4
4
  import 'openai/resources';
package/dist/lib/utils.js CHANGED
@@ -12,7 +12,7 @@
12
12
 
13
13
 
14
14
 
15
- var _chunkTYKJHDNKjs = require('./chunk-TYKJHDNK.js');
15
+ var _chunkIMZJSEAXjs = require('./chunk-IMZJSEAX.js');
16
16
 
17
17
 
18
18
 
@@ -27,4 +27,4 @@ var _chunkTYKJHDNKjs = require('./chunk-TYKJHDNK.js');
27
27
 
28
28
 
29
29
 
30
- exports.getTmpDir = _chunkTYKJHDNKjs.getTmpDir; exports.getTmpFile = _chunkTYKJHDNKjs.getTmpFile; exports.getVersion = _chunkTYKJHDNKjs.getVersion; exports.groupedActionDumpFileExt = _chunkTYKJHDNKjs.groupedActionDumpFileExt; exports.insertScriptBeforeClosingHtml = _chunkTYKJHDNKjs.insertScriptBeforeClosingHtml; exports.overlapped = _chunkTYKJHDNKjs.overlapped; exports.replacerForPageObject = _chunkTYKJHDNKjs.replacerForPageObject; exports.reportHTMLContent = _chunkTYKJHDNKjs.reportHTMLContent; exports.sleep = _chunkTYKJHDNKjs.sleep; exports.stringifyDumpData = _chunkTYKJHDNKjs.stringifyDumpData; exports.uploadTestInfoToServer = _chunkTYKJHDNKjs.uploadTestInfoToServer; exports.writeDumpReport = _chunkTYKJHDNKjs.writeDumpReport; exports.writeLogFile = _chunkTYKJHDNKjs.writeLogFile;
30
+ exports.getTmpDir = _chunkIMZJSEAXjs.getTmpDir; exports.getTmpFile = _chunkIMZJSEAXjs.getTmpFile; exports.getVersion = _chunkIMZJSEAXjs.getVersion; exports.groupedActionDumpFileExt = _chunkIMZJSEAXjs.groupedActionDumpFileExt; exports.insertScriptBeforeClosingHtml = _chunkIMZJSEAXjs.insertScriptBeforeClosingHtml; exports.overlapped = _chunkIMZJSEAXjs.overlapped; exports.replacerForPageObject = _chunkIMZJSEAXjs.replacerForPageObject; exports.reportHTMLContent = _chunkIMZJSEAXjs.reportHTMLContent; exports.sleep = _chunkIMZJSEAXjs.sleep; exports.stringifyDumpData = _chunkIMZJSEAXjs.stringifyDumpData; exports.uploadTestInfoToServer = _chunkIMZJSEAXjs.uploadTestInfoToServer; exports.writeDumpReport = _chunkIMZJSEAXjs.writeDumpReport; exports.writeLogFile = _chunkIMZJSEAXjs.writeLogFile;
@@ -1,9 +1,9 @@
1
- import { ay as StreamingCallback, m as AIUsageInfo, ax as StreamingCodeGenerationOptions, aA as StreamingAIResponse, V as PlanningAction, j as MidsceneYamlFlowItem } from './types-93fa8419.js';
1
+ import { aA as StreamingCallback, m as AIUsageInfo, az as StreamingCodeGenerationOptions, aC as StreamingAIResponse, V as PlanningAction, j as MidsceneYamlFlowItem } from './types-d836fa73.js';
2
2
  import OpenAI from 'openai';
3
3
  import { ChatCompletionMessageParam } from 'openai/resources';
4
4
  export { ChatCompletionMessageParam } from 'openai/resources';
5
- import { b as AIActionType } from './llm-planning-4bc82162.js';
6
- export { a as AiAssert, f as AiExtractElementInfo, A as AiLocateElement, g as AiLocateSection, h as adaptBboxToRect, c as callAiFn, d as describeUserPage, e as elementByPositionWithElementInfo, p as plan } from './llm-planning-4bc82162.js';
5
+ import { b as AIActionType } from './llm-planning-d7096b0d.js';
6
+ export { a as AiAssert, f as AiExtractElementInfo, A as AiLocateElement, g as AiLocateSection, h as adaptBboxToRect, c as callAiFn, d as describeUserPage, e as elementByPositionWithElementInfo, p as plan } from './llm-planning-d7096b0d.js';
7
7
  import { vlLocateMode } from '@midscene/shared/env';
8
8
  import { actionParser } from '@ui-tars/action-parser';
9
9
  import { Size } from '@midscene/shared/types';
@@ -1,7 +1,7 @@
1
- import { E as ExecutionTask, a as ExecutionTaskProgressOptions, b as ExecutionTaskApply, c as ExecutionDump, U as UIContext, I as InsightAction, D as DumpSubscriber, d as InsightTaskInfo, e as InsightOptions, f as DetailedLocateParam, L as LocateResult, g as InsightExtractOption, h as InsightAssertionResponse, A as AIDescribeElementResponse } from './types-93fa8419.js';
2
- export { w as AIAssertionResponse, u as AIDataExtractionResponse, s as AIElementCoordinatesResponse, r as AIElementLocatorResponse, t as AIElementResponse, n as AIResponseFormat, v as AISectionLocatorResponse, q as AISingleElementResponse, o as AISingleElementResponseById, p as AISingleElementResponseByPosition, m as AIUsageInfo, S as AgentAssertOpt, z as AgentDescribeElementAtPointResult, Q as AgentWaitForOpt, a5 as BaseAgentParserOpt, C as CallAIFn, az as CodeGenerationChunk, a4 as Color, H as DumpMeta, N as ElementById, B as EnsureObject, a8 as ExecutionRecorderItem, aq as ExecutionTaskAction, ap as ExecutionTaskActionApply, ab as ExecutionTaskHitBy, ao as ExecutionTaskInsightAssertion, an as ExecutionTaskInsightAssertionApply, am as ExecutionTaskInsightAssertionParam, af as ExecutionTaskInsightDumpLog, ah as ExecutionTaskInsightLocate, ag as ExecutionTaskInsightLocateApply, ae as ExecutionTaskInsightLocateOutput, ad as ExecutionTaskInsightLocateParam, al as ExecutionTaskInsightQuery, ak as ExecutionTaskInsightQueryApply, aj as ExecutionTaskInsightQueryOutput, ai as ExecutionTaskInsightQueryParam, as as ExecutionTaskLog, ar as ExecutionTaskLogApply, au as ExecutionTaskPlanning, at as ExecutionTaskPlanningApply, ac as ExecutionTaskReturn, a9 as ExecutionTaskType, aa as ExecutorContext, aZ as FreeFn, av as GroupedActionDump, J as InsightDump, F as InsightExtractParam, K as LiteUISection, aB as LocateOption, G as LocateResultElement, y as LocateValidatorResult, x as LocatorValidatorOption, b0 as MidsceneYamlConfig, b1 as MidsceneYamlConfigOutput, l as MidsceneYamlConfigResult, j as MidsceneYamlFlowItem, aI as MidsceneYamlFlowItemAIAction, aN as MidsceneYamlFlowItemAIAsk, aJ as MidsceneYamlFlowItemAIAssert, aO as MidsceneYamlFlowItemAIBoolean, aS as MidsceneYamlFlowItemAIHover, aT as MidsceneYamlFlowItemAIInput, aU as MidsceneYamlFlowItemAIKeyboardPress, aP as MidsceneYamlFlowItemAILocate, aL as MidsceneYamlFlowItemAINumber, aK as MidsceneYamlFlowItemAIQuery, k as MidsceneYamlFlowItemAIRightClick, aV as MidsceneYamlFlowItemAIScroll, aM as MidsceneYamlFlowItemAIString, aR as MidsceneYamlFlowItemAITap, aQ as MidsceneYamlFlowItemAIWaitFor, aW as MidsceneYamlFlowItemEvaluateJavaScript, aY as MidsceneYamlFlowItemLogScreenshot, aX as MidsceneYamlFlowItemSleep, M as MidsceneYamlScript, aG as MidsceneYamlScriptAndroidEnv, aH as MidsceneYamlScriptEnv, aE as MidsceneYamlScriptEnvBase, aF as MidsceneYamlScriptWebEnv, i as MidsceneYamlTask, O as OnTaskStartTip, aw as PageType, P as PartialInsightDumpFromSDK, W as PlanningAIResponse, V as PlanningAction, a0 as PlanningActionParamAssert, a2 as PlanningActionParamError, Y as PlanningActionParamHover, _ as PlanningActionParamInputOrKeyPress, Z as PlanningActionParamRightClick, $ as PlanningActionParamScroll, a1 as PlanningActionParamSleep, X as PlanningActionParamTap, a3 as PlanningActionParamWaitFor, T as PlanningLocateParam, a7 as PlaywrightParserOpt, a6 as PuppeteerParserOpt, aC as ReferenceImage, R as ReportDumpWithAttributes, a$ as ScriptPlayerStatusValue, a_ as ScriptPlayerTaskStatus, aA as StreamingAIResponse, ay as StreamingCallback, ax as StreamingCodeGenerationOptions, aD as scrollParam } from './types-93fa8419.js';
3
- import { c as callAiFn } from './llm-planning-4bc82162.js';
4
- export { a as AiAssert, A as AiLocateElement, d as describeUserPage, p as plan } from './llm-planning-4bc82162.js';
1
+ import { E as ExecutionTask, a as ExecutionTaskProgressOptions, b as ExecutionTaskApply, c as ExecutionDump, U as UIContext, I as InsightAction, D as DumpSubscriber, d as InsightTaskInfo, e as InsightOptions, f as DetailedLocateParam, L as LocateResult, g as InsightExtractOption, h as InsightAssertionResponse, A as AIDescribeElementResponse } from './types-d836fa73.js';
2
+ export { w as AIAssertionResponse, u as AIDataExtractionResponse, s as AIElementCoordinatesResponse, r as AIElementLocatorResponse, t as AIElementResponse, n as AIResponseFormat, v as AISectionLocatorResponse, q as AISingleElementResponse, o as AISingleElementResponseById, p as AISingleElementResponseByPosition, m as AIUsageInfo, S as AgentAssertOpt, z as AgentDescribeElementAtPointResult, Q as AgentWaitForOpt, a7 as BaseAgentParserOpt, C as CallAIFn, aB as CodeGenerationChunk, a6 as Color, H as DumpMeta, N as ElementById, B as EnsureObject, aa as ExecutionRecorderItem, as as ExecutionTaskAction, ar as ExecutionTaskActionApply, ad as ExecutionTaskHitBy, aq as ExecutionTaskInsightAssertion, ap as ExecutionTaskInsightAssertionApply, ao as ExecutionTaskInsightAssertionParam, ah as ExecutionTaskInsightDumpLog, aj as ExecutionTaskInsightLocate, ai as ExecutionTaskInsightLocateApply, ag as ExecutionTaskInsightLocateOutput, af as ExecutionTaskInsightLocateParam, an as ExecutionTaskInsightQuery, am as ExecutionTaskInsightQueryApply, al as ExecutionTaskInsightQueryOutput, ak as ExecutionTaskInsightQueryParam, au as ExecutionTaskLog, at as ExecutionTaskLogApply, aw as ExecutionTaskPlanning, av as ExecutionTaskPlanningApply, ae as ExecutionTaskReturn, ab as ExecutionTaskType, ac as ExecutorContext, a$ as FreeFn, ax as GroupedActionDump, J as InsightDump, F as InsightExtractParam, K as LiteUISection, aD as LocateOption, G as LocateResultElement, y as LocateValidatorResult, x as LocatorValidatorOption, b2 as MidsceneYamlConfig, b3 as MidsceneYamlConfigOutput, l as MidsceneYamlConfigResult, j as MidsceneYamlFlowItem, aK as MidsceneYamlFlowItemAIAction, aP as MidsceneYamlFlowItemAIAsk, aL as MidsceneYamlFlowItemAIAssert, aQ as MidsceneYamlFlowItemAIBoolean, aU as MidsceneYamlFlowItemAIHover, aV as MidsceneYamlFlowItemAIInput, aW as MidsceneYamlFlowItemAIKeyboardPress, aR as MidsceneYamlFlowItemAILocate, aN as MidsceneYamlFlowItemAINumber, aM as MidsceneYamlFlowItemAIQuery, k as MidsceneYamlFlowItemAIRightClick, aX as MidsceneYamlFlowItemAIScroll, aO as MidsceneYamlFlowItemAIString, aT as MidsceneYamlFlowItemAITap, aS as MidsceneYamlFlowItemAIWaitFor, aY as MidsceneYamlFlowItemEvaluateJavaScript, a_ as MidsceneYamlFlowItemLogScreenshot, aZ as MidsceneYamlFlowItemSleep, M as MidsceneYamlScript, aI as MidsceneYamlScriptAndroidEnv, aJ as MidsceneYamlScriptEnv, aG as MidsceneYamlScriptEnvBase, aH as MidsceneYamlScriptWebEnv, i as MidsceneYamlTask, O as OnTaskStartTip, ay as PageType, P as PartialInsightDumpFromSDK, W as PlanningAIResponse, V as PlanningAction, a4 as PlanningActionParamAndroidLongPress, a5 as PlanningActionParamAndroidPull, a0 as PlanningActionParamAssert, a2 as PlanningActionParamError, Y as PlanningActionParamHover, _ as PlanningActionParamInputOrKeyPress, Z as PlanningActionParamRightClick, $ as PlanningActionParamScroll, a1 as PlanningActionParamSleep, X as PlanningActionParamTap, a3 as PlanningActionParamWaitFor, T as PlanningLocateParam, a9 as PlaywrightParserOpt, a8 as PuppeteerParserOpt, aE as ReferenceImage, R as ReportDumpWithAttributes, b1 as ScriptPlayerStatusValue, b0 as ScriptPlayerTaskStatus, aC as StreamingAIResponse, aA as StreamingCallback, az as StreamingCodeGenerationOptions, aF as scrollParam } from './types-d836fa73.js';
3
+ import { c as callAiFn } from './llm-planning-d7096b0d.js';
4
+ export { a as AiAssert, A as AiLocateElement, d as describeUserPage, p as plan } from './llm-planning-d7096b0d.js';
5
5
  import { BaseElement, Rect } from '@midscene/shared/types';
6
6
  export { BaseElement, ElementTreeNode, Point, Rect, Size } from '@midscene/shared/types';
7
7
  export { getVersion } from './utils.js';
@@ -1,4 +1,4 @@
1
- import { m as AIUsageInfo, U as UIContext, aC as ReferenceImage, r as AIElementLocatorResponse, N as ElementById, g as InsightExtractOption, u as AIDataExtractionResponse, w as AIAssertionResponse, aw as PageType, W as PlanningAIResponse } from './types-93fa8419.js';
1
+ import { m as AIUsageInfo, U as UIContext, aE as ReferenceImage, r as AIElementLocatorResponse, N as ElementById, g as InsightExtractOption, u as AIDataExtractionResponse, w as AIAssertionResponse, ay as PageType, W as PlanningAIResponse } from './types-d836fa73.js';
2
2
  import { Rect, ElementTreeNode, BaseElement } from '@midscene/shared/types';
3
3
  import { ChatCompletionSystemMessageParam, ChatCompletionUserMessageParam } from 'openai/resources';
4
4
 
@@ -357,7 +357,7 @@ interface PlanningLocateParam extends DetailedLocateParam {
357
357
  }
358
358
  interface PlanningAction<ParamType = any> {
359
359
  thought?: string;
360
- type: 'Locate' | 'Tap' | 'RightClick' | 'Hover' | 'Drag' | 'Input' | 'KeyboardPress' | 'Scroll' | 'Error' | 'ExpectedFalsyCondition' | 'Assert' | 'AssertWithoutThrow' | 'Sleep' | 'Finished' | 'AndroidBackButton' | 'AndroidHomeButton' | 'AndroidRecentAppsButton';
360
+ type: 'Locate' | 'Tap' | 'RightClick' | 'Hover' | 'Drag' | 'Input' | 'KeyboardPress' | 'Scroll' | 'Error' | 'ExpectedFalsyCondition' | 'Assert' | 'AssertWithoutThrow' | 'Sleep' | 'Finished' | 'AndroidBackButton' | 'AndroidHomeButton' | 'AndroidRecentAppsButton' | 'AndroidLongPress' | 'AndroidPull';
361
361
  param: ParamType;
362
362
  locate?: PlanningLocateParam | null;
363
363
  }
@@ -393,6 +393,20 @@ interface PlanningActionParamError {
393
393
  type PlanningActionParamWaitFor = AgentWaitForOpt & {
394
394
  assertion: string;
395
395
  };
396
+ interface PlanningActionParamAndroidLongPress {
397
+ x: number;
398
+ y: number;
399
+ duration?: number;
400
+ }
401
+ interface PlanningActionParamAndroidPull {
402
+ direction: 'up' | 'down';
403
+ startPoint?: {
404
+ x: number;
405
+ y: number;
406
+ };
407
+ distance?: number;
408
+ duration?: number;
409
+ }
396
410
  /**
397
411
  * misc
398
412
  */
@@ -529,4 +543,4 @@ interface StreamingAIResponse {
529
543
  isStreamed: boolean;
530
544
  }
531
545
 
532
- export { type PlanningActionParamScroll as $, type AIDescribeElementResponse as A, type EnsureObject as B, type CallAIFn as C, type DumpSubscriber as D, type ExecutionTask as E, type InsightExtractParam as F, type LocateResultElement as G, type DumpMeta as H, type InsightAction as I, type InsightDump as J, type LiteUISection as K, type LocateResult as L, type MidsceneYamlScript as M, type ElementById as N, type OnTaskStartTip as O, type PartialInsightDumpFromSDK as P, type AgentWaitForOpt as Q, type ReportDumpWithAttributes as R, type AgentAssertOpt as S, type PlanningLocateParam as T, UIContext as U, type PlanningAction as V, type PlanningAIResponse as W, type PlanningActionParamTap as X, type PlanningActionParamHover as Y, type PlanningActionParamRightClick as Z, type PlanningActionParamInputOrKeyPress as _, type ExecutionTaskProgressOptions as a, type ScriptPlayerStatusValue as a$, type PlanningActionParamAssert as a0, type PlanningActionParamSleep as a1, type PlanningActionParamError as a2, type PlanningActionParamWaitFor as a3, type Color as a4, type BaseAgentParserOpt as a5, type PuppeteerParserOpt as a6, type PlaywrightParserOpt as a7, type ExecutionRecorderItem as a8, type ExecutionTaskType as a9, type StreamingAIResponse as aA, type LocateOption as aB, type ReferenceImage as aC, type scrollParam as aD, type MidsceneYamlScriptEnvBase as aE, type MidsceneYamlScriptWebEnv as aF, type MidsceneYamlScriptAndroidEnv as aG, type MidsceneYamlScriptEnv as aH, type MidsceneYamlFlowItemAIAction as aI, type MidsceneYamlFlowItemAIAssert as aJ, type MidsceneYamlFlowItemAIQuery as aK, type MidsceneYamlFlowItemAINumber as aL, type MidsceneYamlFlowItemAIString as aM, type MidsceneYamlFlowItemAIAsk as aN, type MidsceneYamlFlowItemAIBoolean as aO, type MidsceneYamlFlowItemAILocate as aP, type MidsceneYamlFlowItemAIWaitFor as aQ, type MidsceneYamlFlowItemAITap as aR, type MidsceneYamlFlowItemAIHover as aS, type MidsceneYamlFlowItemAIInput as aT, type MidsceneYamlFlowItemAIKeyboardPress as aU, type MidsceneYamlFlowItemAIScroll as aV, type MidsceneYamlFlowItemEvaluateJavaScript as aW, type MidsceneYamlFlowItemSleep as aX, type MidsceneYamlFlowItemLogScreenshot as aY, type FreeFn as aZ, type ScriptPlayerTaskStatus as a_, type ExecutorContext as aa, type ExecutionTaskHitBy as ab, type ExecutionTaskReturn as ac, type ExecutionTaskInsightLocateParam as ad, type ExecutionTaskInsightLocateOutput as ae, type ExecutionTaskInsightDumpLog as af, type ExecutionTaskInsightLocateApply as ag, type ExecutionTaskInsightLocate as ah, type ExecutionTaskInsightQueryParam as ai, type ExecutionTaskInsightQueryOutput as aj, type ExecutionTaskInsightQueryApply as ak, type ExecutionTaskInsightQuery as al, type ExecutionTaskInsightAssertionParam as am, type ExecutionTaskInsightAssertionApply as an, type ExecutionTaskInsightAssertion as ao, type ExecutionTaskActionApply as ap, type ExecutionTaskAction as aq, type ExecutionTaskLogApply as ar, type ExecutionTaskLog as as, type ExecutionTaskPlanningApply as at, type ExecutionTaskPlanning as au, type GroupedActionDump as av, type PageType as aw, type StreamingCodeGenerationOptions as ax, type StreamingCallback as ay, type CodeGenerationChunk as az, type ExecutionTaskApply as b, type MidsceneYamlConfig as b0, type MidsceneYamlConfigOutput as b1, type ExecutionDump as c, type InsightTaskInfo as d, type InsightOptions as e, type DetailedLocateParam as f, type InsightExtractOption as g, type InsightAssertionResponse as h, type MidsceneYamlTask as i, type MidsceneYamlFlowItem as j, type MidsceneYamlFlowItemAIRightClick as k, type MidsceneYamlConfigResult as l, type AIUsageInfo as m, AIResponseFormat as n, type AISingleElementResponseById as o, type AISingleElementResponseByPosition as p, type AISingleElementResponse as q, type AIElementLocatorResponse as r, type AIElementCoordinatesResponse as s, type AIElementResponse as t, type AIDataExtractionResponse as u, type AISectionLocatorResponse as v, type AIAssertionResponse as w, type LocatorValidatorOption as x, type LocateValidatorResult as y, type AgentDescribeElementAtPointResult as z };
546
+ export { type PlanningActionParamScroll as $, type AIDescribeElementResponse as A, type EnsureObject as B, type CallAIFn as C, type DumpSubscriber as D, type ExecutionTask as E, type InsightExtractParam as F, type LocateResultElement as G, type DumpMeta as H, type InsightAction as I, type InsightDump as J, type LiteUISection as K, type LocateResult as L, type MidsceneYamlScript as M, type ElementById as N, type OnTaskStartTip as O, type PartialInsightDumpFromSDK as P, type AgentWaitForOpt as Q, type ReportDumpWithAttributes as R, type AgentAssertOpt as S, type PlanningLocateParam as T, UIContext as U, type PlanningAction as V, type PlanningAIResponse as W, type PlanningActionParamTap as X, type PlanningActionParamHover as Y, type PlanningActionParamRightClick as Z, type PlanningActionParamInputOrKeyPress as _, type ExecutionTaskProgressOptions as a, type FreeFn as a$, type PlanningActionParamAssert as a0, type PlanningActionParamSleep as a1, type PlanningActionParamError as a2, type PlanningActionParamWaitFor as a3, type PlanningActionParamAndroidLongPress as a4, type PlanningActionParamAndroidPull as a5, type Color as a6, type BaseAgentParserOpt as a7, type PuppeteerParserOpt as a8, type PlaywrightParserOpt as a9, type StreamingCallback as aA, type CodeGenerationChunk as aB, type StreamingAIResponse as aC, type LocateOption as aD, type ReferenceImage as aE, type scrollParam as aF, type MidsceneYamlScriptEnvBase as aG, type MidsceneYamlScriptWebEnv as aH, type MidsceneYamlScriptAndroidEnv as aI, type MidsceneYamlScriptEnv as aJ, type MidsceneYamlFlowItemAIAction as aK, type MidsceneYamlFlowItemAIAssert as aL, type MidsceneYamlFlowItemAIQuery as aM, type MidsceneYamlFlowItemAINumber as aN, type MidsceneYamlFlowItemAIString as aO, type MidsceneYamlFlowItemAIAsk as aP, type MidsceneYamlFlowItemAIBoolean as aQ, type MidsceneYamlFlowItemAILocate as aR, type MidsceneYamlFlowItemAIWaitFor as aS, type MidsceneYamlFlowItemAITap as aT, type MidsceneYamlFlowItemAIHover as aU, type MidsceneYamlFlowItemAIInput as aV, type MidsceneYamlFlowItemAIKeyboardPress as aW, type MidsceneYamlFlowItemAIScroll as aX, type MidsceneYamlFlowItemEvaluateJavaScript as aY, type MidsceneYamlFlowItemSleep as aZ, type MidsceneYamlFlowItemLogScreenshot as a_, type ExecutionRecorderItem as aa, type ExecutionTaskType as ab, type ExecutorContext as ac, type ExecutionTaskHitBy as ad, type ExecutionTaskReturn as ae, type ExecutionTaskInsightLocateParam as af, type ExecutionTaskInsightLocateOutput as ag, type ExecutionTaskInsightDumpLog as ah, type ExecutionTaskInsightLocateApply as ai, type ExecutionTaskInsightLocate as aj, type ExecutionTaskInsightQueryParam as ak, type ExecutionTaskInsightQueryOutput as al, type ExecutionTaskInsightQueryApply as am, type ExecutionTaskInsightQuery as an, type ExecutionTaskInsightAssertionParam as ao, type ExecutionTaskInsightAssertionApply as ap, type ExecutionTaskInsightAssertion as aq, type ExecutionTaskActionApply as ar, type ExecutionTaskAction as as, type ExecutionTaskLogApply as at, type ExecutionTaskLog as au, type ExecutionTaskPlanningApply as av, type ExecutionTaskPlanning as aw, type GroupedActionDump as ax, type PageType as ay, type StreamingCodeGenerationOptions as az, type ExecutionTaskApply as b, type ScriptPlayerTaskStatus as b0, type ScriptPlayerStatusValue as b1, type MidsceneYamlConfig as b2, type MidsceneYamlConfigOutput as b3, type ExecutionDump as c, type InsightTaskInfo as d, type InsightOptions as e, type DetailedLocateParam as f, type InsightExtractOption as g, type InsightAssertionResponse as h, type MidsceneYamlTask as i, type MidsceneYamlFlowItem as j, type MidsceneYamlFlowItemAIRightClick as k, type MidsceneYamlConfigResult as l, type AIUsageInfo as m, AIResponseFormat as n, type AISingleElementResponseById as o, type AISingleElementResponseByPosition as p, type AISingleElementResponse as q, type AIElementLocatorResponse as r, type AIElementCoordinatesResponse as s, type AIElementResponse as t, type AIDataExtractionResponse as u, type AISectionLocatorResponse as v, type AIAssertionResponse as w, type LocatorValidatorOption as x, type LocateValidatorResult as y, type AgentDescribeElementAtPointResult as z };
@@ -1,4 +1,4 @@
1
- import { R as ReportDumpWithAttributes } from './types-93fa8419.js';
1
+ import { R as ReportDumpWithAttributes } from './types-d836fa73.js';
2
2
  import { Rect } from '@midscene/shared/types';
3
3
  import '@midscene/shared/constants';
4
4
  import 'openai/resources';
package/package.json CHANGED
@@ -1,7 +1,7 @@
1
1
  {
2
2
  "name": "@midscene/core",
3
3
  "description": "Automate browser actions, extract data, and perform assertions using AI. It offers JavaScript SDK, Chrome extension, and support for scripting in YAML. See https://midscenejs.com/ for details.",
4
- "version": "0.24.2-beta-20250731102409.0",
4
+ "version": "0.24.2-beta-20250731132300.0",
5
5
  "repository": "https://github.com/web-infra-dev/midscene",
6
6
  "homepage": "https://midscenejs.com/",
7
7
  "jsnext:source": "./src/index.ts",
@@ -44,8 +44,8 @@
44
44
  "langsmith": "0.3.7",
45
45
  "openai": "4.81.0",
46
46
  "socks-proxy-agent": "8.0.4",
47
- "@midscene/recorder": "0.24.2-beta-20250731102409.0",
48
- "@midscene/shared": "0.24.2-beta-20250731102409.0"
47
+ "@midscene/recorder": "0.24.2-beta-20250731132300.0",
48
+ "@midscene/shared": "0.24.2-beta-20250731132300.0"
49
49
  },
50
50
  "devDependencies": {
51
51
  "@modern-js/module-tools": "2.60.6",