v0.1.2: 全项目代码审查修复 + 子代理编排器 + 上下文压缩 + 并行工具执行
后端修复: Ollama adapter 移除死代码/修复超时硬编码/iteration硬编码/pullModel无超时/流异常断开补发DONE; SSE解析器支持非字符串arguments; Sandbox fail-closed安全加固; IPC移除未使用变量 新增功能: TaskOrchestrator子任务编排器; DelegateTaskTool委派工具; Agent Loop并行工具执行; 上下文自动压缩; 记忆检索注入System Prompt 前端修复: useAgentStream text_delta thought累积bug; 工具调用状态正确流转; 修复重复stateChange事件; TraceStep.thought正确填充
This commit is contained in:
@@ -4,12 +4,14 @@
|
||||
* 三阶段回退策略:
|
||||
* Phase 1: HTTP 抓取(UA 轮换 + 反爬请求头 + 指数退避重试 + 拦截检测)
|
||||
* Phase 2: 内容过短自动升级(< 200 字符 → 浏览器渲染)
|
||||
* Phase 3: 浏览器回退(隐藏 BrowserWindow + JS 渲染 + 内容提取)
|
||||
* Phase 3: 浏览器回退(共享 BrowserWindowManager + JS 渲染 + 内容提取)
|
||||
*
|
||||
* 浏览器回退使用与 web_browser 相同的 BrowserWindowManager 单例,
|
||||
* 避免创建多个独立浏览器窗口,支持窗口复用。
|
||||
*
|
||||
* @see docs/Agent网络工具通用设计-v2.md — 第 3 章 web_fetch 抓取设计
|
||||
*/
|
||||
|
||||
import { BrowserWindow } from 'electron';
|
||||
import type { IMetonaTool, ToolExecutionContext } from '../../types/metona-tool';
|
||||
import type { MetonaToolDef } from '../../../harness/types';
|
||||
import { MetonaToolCategory, MetonaRiskLevel } from '../../../harness/types';
|
||||
@@ -22,6 +24,7 @@ import {
|
||||
readBodyWithLimit,
|
||||
logTool,
|
||||
} from './network-utils';
|
||||
import { getBrowserManager } from './browser';
|
||||
|
||||
// ===== 跳过重试的状态码 =====
|
||||
|
||||
@@ -57,6 +60,13 @@ export class WebFetchTool implements IMetonaTool {
|
||||
return { url, content: '', success: false, error: 'URL must start with http:// or https://' };
|
||||
}
|
||||
|
||||
// 先查缓存(HTTP 和浏览器阶段共享同一缓存)
|
||||
const cached = fetchCache.get(url);
|
||||
if (cached) {
|
||||
logTool('web_fetch', `Cache hit: ${url}`);
|
||||
return { url, content: cached, success: true, method: 'cache', length: cached.length };
|
||||
}
|
||||
|
||||
logTool('web_fetch', `Fetching: ${url}`);
|
||||
|
||||
// ===== Phase 1: HTTP 抓取 =====
|
||||
@@ -71,6 +81,8 @@ export class WebFetchTool implements IMetonaTool {
|
||||
return this.buildSuccess(url, browserResult, 'browser');
|
||||
}
|
||||
}
|
||||
// 写入缓存
|
||||
fetchCache.set(url, phase1Result.text);
|
||||
return this.buildSuccess(url, phase1Result.text, 'http');
|
||||
}
|
||||
|
||||
@@ -144,7 +156,7 @@ export class WebFetchTool implements IMetonaTool {
|
||||
return { success: false, text: '', intercepted: false, reason: 'All retries exhausted' };
|
||||
}
|
||||
|
||||
// ===== Phase 2/3: 浏览器回退 =====
|
||||
// ===== Phase 2/3: 浏览器回退(使用共享 BrowserWindowManager 单例) =====
|
||||
|
||||
private async browserFetch(url: string): Promise<string | null> {
|
||||
// 查缓存
|
||||
@@ -154,59 +166,50 @@ export class WebFetchTool implements IMetonaTool {
|
||||
return cached;
|
||||
}
|
||||
|
||||
let win: BrowserWindow | null = null;
|
||||
let cleanup: (() => void) | null = null;
|
||||
|
||||
try {
|
||||
win = new BrowserWindow({
|
||||
width: 1280,
|
||||
height: 800,
|
||||
show: false,
|
||||
webPreferences: {
|
||||
nodeIntegration: false,
|
||||
contextIsolation: true,
|
||||
sandbox: true,
|
||||
webSecurity: false, // 允许跨域(截图需要)
|
||||
},
|
||||
});
|
||||
const manager = getBrowserManager();
|
||||
|
||||
cleanup = () => {
|
||||
try {
|
||||
if (win && !win.isDestroyed()) win.close();
|
||||
} catch {
|
||||
// 忽略关闭错误
|
||||
}
|
||||
};
|
||||
|
||||
await this.loadURLWithTimeout(win, url, 30_000);
|
||||
// 通过 manager 打开 URL(复用已打开的同 URL 窗口,避免重复创建)
|
||||
await manager.open({ url });
|
||||
|
||||
// 等待 JS 渲染
|
||||
await this.sleep(2_500);
|
||||
|
||||
// 提取页面正文
|
||||
const text = await win.webContents.executeJavaScript(`
|
||||
const text = await manager.evaluate(`
|
||||
(function() {
|
||||
var clone = document.body.cloneNode(true);
|
||||
var noise = clone.querySelectorAll('script, style, noscript, nav, header, footer, aside, iframe, svg');
|
||||
noise.forEach(function(el) { el.remove(); });
|
||||
return clone.innerText || '';
|
||||
})();
|
||||
`, true);
|
||||
`) as string;
|
||||
|
||||
if (text && text.trim().length >= 80) {
|
||||
// 拦截检测(浏览器渲染后仍可能是验证码挑战页)
|
||||
if (isInterceptedPage(text)) {
|
||||
logTool('web_fetch', `Browser fetch detected intercepted page: ${url}`);
|
||||
return null;
|
||||
}
|
||||
|
||||
// 内容大小限制(与 HTTP 阶段一致,防止超大页面耗尽上下文)
|
||||
const MAX_BROWSER_TEXT = 500_000; // 500K chars
|
||||
const safeText = text.length > MAX_BROWSER_TEXT
|
||||
? text.slice(0, MAX_BROWSER_TEXT) + '\n\n[... content truncated ...]'
|
||||
: text;
|
||||
|
||||
// 写缓存
|
||||
fetchCache.set(url, text);
|
||||
logTool('web_fetch', `Browser fetch success: ${text.length} chars`);
|
||||
return text;
|
||||
fetchCache.set(url, safeText);
|
||||
logTool('web_fetch', `Browser fetch success: ${safeText.length} chars`);
|
||||
return safeText;
|
||||
}
|
||||
|
||||
return null;
|
||||
} catch (err) {
|
||||
logTool('web_fetch', `Browser fetch failed: ${(err as Error).message}`);
|
||||
return null;
|
||||
} finally {
|
||||
if (cleanup) cleanup();
|
||||
}
|
||||
// 注意:不关闭窗口 — manager 是单例,窗口由 web_browser 或 cleanupBrowser 管理
|
||||
}
|
||||
|
||||
// ===== 辅助方法 =====
|
||||
@@ -224,17 +227,4 @@ export class WebFetchTool implements IMetonaTool {
|
||||
private sleep(ms: number): Promise<void> {
|
||||
return new Promise((resolve) => setTimeout(resolve, ms));
|
||||
}
|
||||
|
||||
/** 带超时的 loadURL(Electron 原生不支持 timeout 选项) */
|
||||
private async loadURLWithTimeout(win: BrowserWindow, url: string, timeoutMs: number): Promise<void> {
|
||||
let timer: NodeJS.Timeout | null = null;
|
||||
const timeoutPromise = new Promise<never>((_, reject) => {
|
||||
timer = setTimeout(() => reject(new Error(`Page load timeout after ${timeoutMs}ms: ${url}`)), timeoutMs);
|
||||
});
|
||||
try {
|
||||
await Promise.race([win.loadURL(url), timeoutPromise]);
|
||||
} finally {
|
||||
if (timer) clearTimeout(timer);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user