page-agent 0.0.5 → 0.0.7

package/README.md

@@ -67,16 +67,17 @@ await agent.execute('Click the login button')
 
 ## 🏗️ Structure
 
-PageAgent follows a clean, modular architecture:
+PageAgent adopts a simplified monorepo structure:
 
 ```
-src/
-├── PageAgent.ts          # Agent main loop
-├── dom/                  # DOM processing
-├── tools/                # Agent tools
-├── ui/                   # UI components & panels
-├── llms/                 # LLM integration layer
-└── utils/                # Event bus & utilities
+packages/
+├── page-agent/          # AI agent (npm: page-agent)
+│   ├── PageAgent        # Agent main loop
+│   ├── tools/           # LLM tool definitions
+│   ├── ui/              # UI components & panels
+│   └── llms/            # LLM integration layer
+├── page-controller/     # DOM operations (npm: @page-agent/page-controller)
+└── website/             # Demo & Documentation site
 ```
 
 ## 🤝 Contributing

package/dist/lib/PageAgent.d.ts

@@ -1,4 +1,6 @@
 import { Motion } from 'ai-motion';
+import { PageController } from '@page-agent/page-controller';
+import { PageControllerConfig } from '@page-agent/page-controller';
 import { z } from 'zod';
 
 export declare interface AgentBrain {
@@ -89,31 +91,6 @@ declare type DeepStringify<T> = {
     [K in keyof T]: T[K] extends string ? string : T[K] extends object ? DeepStringify<T[K]> : T[K];
 };
 
-declare interface DomConfig {
-    interactiveBlacklist?: (Element | (() => Element))[];
-    interactiveWhitelist?: (Element | (() => Element))[];
-    include_attributes?: string[];
-    highlightOpacity?: number;
-    highlightLabelOpacity?: number;
-}
-
-declare type DomNode = TextDomNode | ElementDomNode | InteractiveElementDomNode;
-
-declare interface ElementDomNode {
-    tagName: string;
-    attributes?: Record<string, string>;
-    xpath?: string;
-    children?: string[];
-    isVisible?: boolean;
-    isTopElement?: boolean;
-    isInViewport?: boolean;
-    isNew?: boolean;
-    isInteractive?: false;
-    highlightIndex?: number;
-    extra?: Record<string, any>;
-    [key: string]: unknown;
-}
-
 declare const enUS: {
     readonly ui: {
         readonly panel: {
@@ -196,11 +173,6 @@ export declare interface ExecutionResult {
     history: AgentHistory[];
 }
 
-declare interface FlatDomTree {
-    rootId: string;
-    map: Record<string, DomNode>;
-}
-
 declare class I18n {
     private language;
     private translations;
@@ -211,23 +183,6 @@ declare class I18n {
     getLanguage(): SupportedLanguage;
 }
 
-declare interface InteractiveElementDomNode {
-    tagName: string;
-    attributes?: Record<string, string>;
-    xpath?: string;
-    children?: string[];
-    isVisible?: boolean;
-    isTopElement?: boolean;
-    isInViewport?: boolean;
-    isInteractive: true;
-    highlightIndex: number;
-    /**
-     * 可交互元素的 dom 引用
-     */
-    ref: HTMLElement;
-    [key: string]: unknown;
-}
-
 declare interface LLMConfig {
     baseURL?: string;
     apiKey?: string;
@@ -370,19 +325,8 @@ export declare class PageAgent extends EventTarget {
     disposed: boolean;
     task: string;
     taskId: string;
-    /** Corresponds to eval_page in browser-use */
-    flatTree: FlatDomTree | null;
-    /**
-     * All highlighted index-mapped interactive elements
-     * Corresponds to DOMState.selector_map in browser-use
-     */
-    selectorMap: Map<number, InteractiveElementDomNode>;
-    /** highlight index -> element text */
-    elementTextMap: Map<number, string>;
-    /** Corresponds to clickable_elements_to_string in browser-use */
-    simplifiedHTML: string;
-    /** last time the tree was updated */
-    lastTimeUpdate: number;
+    /** PageController for DOM operations */
+    pageController: PageController;
     /** Fullscreen mask */
     mask: SimulatorMask;
     /** History records */
@@ -395,7 +339,7 @@ export declare class PageAgent extends EventTarget {
     dispose(reason?: string): void;
 }
 
-export declare type PageAgentConfig = LLMConfig & AgentConfig & DomConfig;
+export declare type PageAgentConfig = LLMConfig & AgentConfig & PageControllerConfig;
 
 /**
  * Event mapping definitions
@@ -477,13 +421,6 @@ declare interface Step {
 
 declare type SupportedLanguage = keyof typeof locales;
 
-declare interface TextDomNode {
-    type: 'TEXT_NODE';
-    text: string;
-    isVisible: boolean;
-    [key: string]: unknown;
-}
-
 export declare function tool<TParams>(options: PageAgentTool<TParams>): PageAgentTool<TParams>;
 
 /**