ai-agent-book 精选快照（<2MB 代码与文档，来自 github.com/bojieli/ai-agent-book）

2026-08-20 13:12:50 +00:00
commit b119135836
10275 changed files with 3284984 additions and 0 deletions
@@ -0,0 +1,56 @@
+---
+title: "Basics"
+description: "Low-level Playwright-like browser automation with direct and full CDP control and precise element interactions"
+icon: "code"
+mode: "wide"
+---
+
+## Core Architecture
+
+```mermaid
+graph TD
+    A[Browser] --> B[Page]
+    B --> C[Element]
+    B --> D[Mouse]
+    B --> E[AI Features]
+    C --> F[DOM Interactions]
+    D --> G[Coordinate Operations]
+    E --> H[LLM Integration]
+```
+
+### Core Classes
+
+- **Browser** (alias: **BrowserSession**): Main session manager
+- **Page**: Represents a browser tab/iframe
+- **Element**: Individual DOM element operations
+- **Mouse**: Coordinate-based mouse operations
+
+## Basic Usage
+
+```python
+from browser_use import Browser, Agent
+from browser_use.llm.openai import ChatOpenAI
+
+async def main():
+    llm = ChatOpenAI(api_key="your-api-key")
+    browser = Browser()
+    await browser.start()
+
+    # 1. Actor: Precise navigation and element interactions
+    page = await browser.new_page("https://github.com/login")
+    email_input = await page.must_get_element_by_prompt("username field", llm=llm)
+    await email_input.fill("your-username")
+
+    # 2. Agent: AI-driven complex tasks
+    agent = Agent(browser=browser, llm=llm)
+    await agent.run("Complete login and navigate to my repositories")
+
+    await browser.stop()
+```
+
+## Important Notes
+
+- **Not Playwright**: Actor is built on CDP, not Playwright. The API resembles Playwright as much as possible for easy migration, but is sorta subset.
+- **Immediate Returns**: `get_elements_by_css_selector()` doesn't wait for visibility
+- **Manual Timing**: You handle navigation timing and waiting
+- **JavaScript Format**: `evaluate()` requires arrow function format: `() => {}`