Merge branch 'main' into feat/add-issue-templates

Author: ycjcl868

apps/agent-tars/src/renderer/src/agent/AgentFlow.ts

@@ -137,7 +137,7 @@ export class AgentFlow {
         async (event: GlobalEvent) => {
           switch (event.type) {
             case 'user-interrupt':
-              await this.eventManager.addUserInteruptionInput(event.text);
+              await this.eventManager.addUserInterruptionInput(event.text);
               this.interruptController.abort();
               await chatUtils.updateMessage(
                 ChatMessageUtil.assistantOmegaMessage({

apps/agent-tars/src/renderer/src/agent/EventManager.ts

@@ -186,8 +186,8 @@ export class EventManager {
   /**
    * Add a user interrupt instruction event
    */
-  public async addUserInteruptionInput(text: string): Promise<EventItem> {
-    return this.addEvent(EventType.UserInteruption, { text });
+  public async addUserInterruptionInput(text: string): Promise<EventItem> {
+    return this.addEvent(EventType.UserInterruption, { text });
   }
 
   /**
@@ -409,12 +409,12 @@ export class EventManager {
           },
         };
 
-      case EventType.UserInteruption:
+      case EventType.UserInterruption:
         return {
           ...base,
           content: {
             text: (
-              event.content as EventContentDescriptor[EventType.UserInteruption]
+              event.content as EventContentDescriptor[EventType.UserInterruption]
             ).text,
           },
         };

apps/agent-tars/src/renderer/src/agent/mockEvents.ts

@@ -1567,7 +1567,7 @@ export const events = [
   },
   {
     id: 'ee4942bb-1a6e-4caf-86a7-d7b55c8f0134',
-    type: 'user-interuption',
+    type: 'user-interruption',
     content: {
       text: '直接到最后一步',
     },

apps/agent-tars/src/renderer/src/components/AgentFlowMessage/EventRenderer.tsx

@@ -4,7 +4,7 @@ import { ToolUsed } from './events/ToolUsed';
 import { AgentStatus } from './events/AgentStatus';
 import { ChatText } from './events/ChatText';
 // import { Observation } from './events/Observation';
-import { UserInteruption } from './events/UserInteruption';
+import { UserInterruption } from './events/UserInterruption';
 import { AgentFlowEnd } from './events/End';
 
 export function EventRenderer({
@@ -20,7 +20,7 @@ export function EventRenderer({
     [EventType.AgentStatus]: AgentStatus,
     [EventType.ChatText]: ChatText,
     // [EventType.Observation]: Observation,
-    [EventType.UserInteruption]: UserInteruption,
+    [EventType.UserInterruption]: UserInterruption,
     [EventType.End]: AgentFlowEnd,
   };
 

apps/agent-tars/src/renderer/src/components/AgentFlowMessage/events/UserInteruption.tsx

@@ -6,7 +6,7 @@ import relativeTime from 'dayjs/plugin/relativeTime';
 
 dayjs.extend(relativeTime);
 
-export function UserInteruption({
+export function UserInterruption({
   event,
   isLastEvent,
 }: {

apps/agent-tars/src/renderer/src/type/event.ts

@@ -10,7 +10,7 @@ export enum EventType {
   ChatText = 'chat-text',
   Observation = 'observation',
   NewPlanStep = 'new-plan-step',
-  UserInteruption = 'user-interuption',
+  UserInterruption = 'user-interruption',
   End = 'end',
 }
 
@@ -52,7 +52,7 @@ export interface EventContentDescriptor {
   };
   // When tools return the result, it will be displayed as an observation
   [EventType.Observation]: any;
-  [EventType.UserInteruption]: {
+  [EventType.UserInterruption]: {
     text: string;
   };
   [EventType.End]: {

packages/agent-infra/mcp-servers/commands/TODO.md

@@ -7,7 +7,7 @@
         - And, don't put any constraints on the memory other than a list of strings. Claude can do all that on his own (i.e. a line with `pythoh3` alone should be sufficient, in fact I could do some testing with that alone and see how Claude does)
     - Another example would be `uname` on a Windows machine. 
     - These different scenarios aren't necessarily mission critical to fix but they offer an opportunity to improve the experience.
-    - The solution may not be generalizable too and might be specific to commmand that fails, i.e. if I pass the OS name in the ListTools response, that will likely fix that issue.
+    - The solution may not be generalizable too and might be specific to command that fails, i.e. if I pass the OS name in the ListTools response, that will likely fix that issue.
     - i.e. routinely `python` is used and then `python3`
     - Or, should I have some static mappings of common commands that fail and when they do, use the fallback? And find a way to tell the LLM? Or,
     - Or would some sort of command lookup mechanism be useful? i.e. python3 instead of python

packages/agent-infra/search/browser-search/examples/github-search.ts

@@ -6,7 +6,7 @@ import { ConsoleLogger } from '@agent-infra/logger';
 import { BrowserSearch } from '../src';
 
 /**
- * We commmented github engine temporarily,
+ * We commented github engine temporarily,
  * since it does not work,
  * we'll figure out later.
  */

packages/ui-tars/sdk/src/Model.ts

@@ -13,7 +13,7 @@ import { useContext } from './context/useContext';
 import { Model, type InvokeParams, type InvokeOutput } from './types';
 
 import { preprocessResizeImage, convertToOpenAIMessages } from './utils';
-import { DEFUALT_FACTORS, MAX_PIXELS } from './constants';
+import { DEFAULT_FACTORS, MAX_PIXELS } from './constants';
 
 type OpenAIChatCompletionCreateParams = Omit<ClientOptions, 'maxRetries'> &
   Pick<
@@ -31,7 +31,7 @@ export class UITarsModel extends Model {
 
   /** [widthFactor, heightFactor] */
   get factors(): [number, number] {
-    return DEFUALT_FACTORS;
+    return DEFAULT_FACTORS;
   }
 
   get modelName(): string {

packages/ui-tars/sdk/src/constants.ts

@@ -7,7 +7,7 @@ import type { AgentContext } from './types';
 export type Factors = [number, number];
 
 export const MAX_SNAPSHOT_ERR_CNT = 10;
-export const DEFUALT_FACTORS: Factors = [1000, 1000];
+export const DEFAULT_FACTORS: Factors = [1000, 1000];
 export const MAX_PIXELS = 1350 * 28 * 28;
 export const SYSTEM_PROMPT = `You are a GUI agent. You are given a task and your action history, with screenshots. You need to perform the next action to complete the task.
 
@@ -37,7 +37,7 @@ call_user() # Submit the task and call the user when the task is unsolvable, or
 
 export const DEFAULT_CONTEXT = {
   logger: console,
-  factors: DEFUALT_FACTORS,
+  factors: DEFAULT_FACTORS,
   systemPrompt: SYSTEM_PROMPT,
 } satisfies Partial<AgentContext>;
 

packages/ui-tars/sdk/src/utils.test.ts

@@ -6,19 +6,19 @@
 import { describe, expect, it } from 'vitest';
 
 import { parseBoxToScreenCoords } from './utils';
-import { DEFUALT_FACTORS } from './constants';
+import { DEFAULT_FACTORS } from './constants';
 
 describe('parseBoxToScreenCoords', () => {
   it('should correctly parse single point coordinates', () => {
     const result = parseBoxToScreenCoords({
       boxStr: '[0.5,0.5]',
       screenWidth: 1000,
       screenHeight: 800,
-      factors: DEFUALT_FACTORS,
+      factors: DEFAULT_FACTORS,
     });
     expect(result).toEqual({
-      x: Math.round(0.5 * 1000 * DEFUALT_FACTORS[0]) / DEFUALT_FACTORS[0],
-      y: Math.round(0.5 * 800 * DEFUALT_FACTORS[1]) / DEFUALT_FACTORS[1],
+      x: Math.round(0.5 * 1000 * DEFAULT_FACTORS[0]) / DEFAULT_FACTORS[0],
+      y: Math.round(0.5 * 800 * DEFAULT_FACTORS[1]) / DEFAULT_FACTORS[1],
     });
   });
 
@@ -29,8 +29,8 @@ describe('parseBoxToScreenCoords', () => {
       screenHeight: 800,
     });
     expect(result).toEqual({
-      x: Math.round(0.5 * 1000 * DEFUALT_FACTORS[0]) / DEFUALT_FACTORS[0],
-      y: Math.round(0.5 * 800 * DEFUALT_FACTORS[1]) / DEFUALT_FACTORS[1],
+      x: Math.round(0.5 * 1000 * DEFAULT_FACTORS[0]) / DEFAULT_FACTORS[0],
+      y: Math.round(0.5 * 800 * DEFAULT_FACTORS[1]) / DEFAULT_FACTORS[1],
     });
   });
 
@@ -39,11 +39,11 @@ describe('parseBoxToScreenCoords', () => {
       boxStr: '[0.2,0.3,0.4,0.5]',
       screenWidth: 1000,
       screenHeight: 800,
-      factors: DEFUALT_FACTORS,
+      factors: DEFAULT_FACTORS,
     });
     expect(result).toEqual({
-      x: Math.round(0.3 * 1000 * DEFUALT_FACTORS[0]) / DEFUALT_FACTORS[0], // (0.2 + 0.4) / 2 = 0.3
-      y: Math.round(0.4 * 800 * DEFUALT_FACTORS[1]) / DEFUALT_FACTORS[1], // (0.3 + 0.5) / 2 = 0.4
+      x: Math.round(0.3 * 1000 * DEFAULT_FACTORS[0]) / DEFAULT_FACTORS[0], // (0.2 + 0.4) / 2 = 0.3
+      y: Math.round(0.4 * 800 * DEFAULT_FACTORS[1]) / DEFAULT_FACTORS[1], // (0.3 + 0.5) / 2 = 0.4
     });
   });
 
@@ -52,11 +52,11 @@ describe('parseBoxToScreenCoords', () => {
       boxStr: '[ 0.5 , 0.5 ]',
       screenWidth: 1000,
       screenHeight: 800,
-      factors: DEFUALT_FACTORS,
+      factors: DEFAULT_FACTORS,
     });
     expect(result).toEqual({
-      x: Math.round(0.5 * 1000 * DEFUALT_FACTORS[0]) / DEFUALT_FACTORS[0],
-      y: Math.round(0.5 * 800 * DEFUALT_FACTORS[1]) / DEFUALT_FACTORS[1],
+      x: Math.round(0.5 * 1000 * DEFAULT_FACTORS[0]) / DEFAULT_FACTORS[0],
+      y: Math.round(0.5 * 800 * DEFAULT_FACTORS[1]) / DEFAULT_FACTORS[1],
     });
   });
 
@@ -65,11 +65,11 @@ describe('parseBoxToScreenCoords', () => {
       boxStr: '[1,1,2,2]',
       screenWidth: 1000,
       screenHeight: 800,
-      factors: DEFUALT_FACTORS,
+      factors: DEFAULT_FACTORS,
     });
     expect(result).toEqual({
-      x: Math.round(1.5 * 1000 * DEFUALT_FACTORS[0]) / DEFUALT_FACTORS[0],
-      y: Math.round(1.5 * 800 * DEFUALT_FACTORS[1]) / DEFUALT_FACTORS[1],
+      x: Math.round(1.5 * 1000 * DEFAULT_FACTORS[0]) / DEFAULT_FACTORS[0],
+      y: Math.round(1.5 * 800 * DEFAULT_FACTORS[1]) / DEFAULT_FACTORS[1],
     });
   });
 
@@ -78,7 +78,7 @@ describe('parseBoxToScreenCoords', () => {
       boxStr: '',
       screenWidth: 1000,
       screenHeight: 800,
-      factors: DEFUALT_FACTORS,
+      factors: DEFAULT_FACTORS,
     });
     expect(result).toEqual({
       x: null,

packages/ui-tars/sdk/src/utils.ts

@@ -7,7 +7,7 @@ import { ChatCompletionMessageParam } from 'openai/resources/chat/completions';
 
 import { IMAGE_PLACEHOLDER, MAX_IMAGE_LENGTH } from '@ui-tars/shared/constants';
 import { Conversation, Message } from '@ui-tars/shared/types';
-import { DEFUALT_FACTORS, type Factors } from './constants';
+import { DEFAULT_FACTORS, type Factors } from './constants';
 
 /**
  * Parse box string to screen coordinates
@@ -24,7 +24,7 @@ export const parseBoxToScreenCoords = ({
   boxStr,
   screenWidth,
   screenHeight,
-  factors = DEFUALT_FACTORS,
+  factors = DEFAULT_FACTORS,
 }: {
   boxStr: string;
   screenWidth: number;

packages/ui-tars/sdk/tests/GUIAgent.test.ts

@@ -12,7 +12,7 @@ import { GUIAgentData, StatusEnum } from '../src';
 import { IMAGE_PLACEHOLDER } from '@ui-tars/shared/constants';
 import { UITarsModel } from '../src/Model';
 import { mockOpenAIResponse } from './testKits/index';
-import { DEFUALT_FACTORS } from '../src/constants';
+import { DEFAULT_FACTORS } from '../src/constants';
 
 const getContext = vi.fn();
 vi.mock('openai', () => ({
@@ -80,7 +80,7 @@ describe('GUIAgent', () => {
 
     expect(operator.execute).toBeCalledTimes(1);
     expect(operator.execute.mock.calls[0][0]).toEqual({
-      factors: DEFUALT_FACTORS,
+      factors: DEFAULT_FACTORS,
       parsedPrediction: {
         action_inputs: {
           start_box: '[0.072,0.646,0.072,0.646]',
@@ -387,7 +387,7 @@ describe('GUIAgent', () => {
 
     expect(operator.execute).toBeCalledTimes(2);
     expect(operator.execute.mock.calls[0][0]).toEqual({
-      factors: DEFUALT_FACTORS,
+      factors: DEFAULT_FACTORS,
       parsedPrediction: {
         action_inputs: {
           start_box: '[0.072,0.646,0.072,0.646]',