Skip to content

refactor(gui-agent): refactor composable gui agent with improved stability and error handling #1556

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
cjraft merged 6 commits into from
Sep 25, 2025
Merged

refactor(gui-agent): refactor composable gui agent with improved stability and error handling #1556

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
37ee17f
refactor(gui-agent): update aio hybrid operator with refactored api
ZhaoHeh Sep 17, 2025
f6e1c9a
refactor(o-gui-agent): updated with refactored action parser and aio …
ZhaoHeh Sep 19, 2025
db7d495
refactor(o-gui-agent): input screenshot on AfterToolCall hook
ZhaoHeh Sep 21, 2025
b46cdb8
chore(gui-agent): fix point undefined type error
ZhaoHeh Sep 24, 2025
5419387
feat(aio-hybrid-operator): improve connection stability and error rec…
ZhaoHeh Sep 24, 2025
dadd4dd
chore(aio-hybrid-operator): delete utils test file
ZhaoHeh Sep 24, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 3 additions & 7 deletions multimodal/gui-agent/operator-aio/package.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,11 +10,8 @@
"url": "https://github.com/bytedance/UI-TARS-desktop/issues"
},
"keywords": [
"AI",
"Core",
"SDK",
"Operator",
"UI-TARS",
"Operator",
"AIO"
],
"main": "./dist/index.js",
Expand All @@ -36,12 +33,11 @@
"registry": "https://registry.npmjs.org"
},
"dependencies": {
"@gui-agent/shared": "workspace:*",
"@agent-infra/browser": "0.1.1",
"@agent-infra/logger": "0.0.2-beta.2",
"@agent-infra/sandbox": "0.0.6",
"@agent-infra/media-utils": "0.1.5",
"@ui-tars/sdk": "1.2.3",
"@ui-tars/shared": "1.2.3"
"@agent-infra/media-utils": "0.1.5"
},
"devDependencies": {
"@rslib/core": "0.10.0",
Expand Down
10 changes: 9 additions & 1 deletion multimodal/gui-agent/operator-aio/src/AIOBrowser.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,6 +41,8 @@ export class AIOBrowser {
*/
private logger: Logger;

private launchOptions?: LaunchOptions;

/**
* Creates an instance of AIOBrowser
* @param {AIOBrowserOptions} options - Configuration options
Expand All @@ -64,6 +66,7 @@ export class AIOBrowser {
try {
await this.browser.launch(options);
this.logger.success('Browser launched successfully');
this.launchOptions = options;
} catch (error) {
this.logger.error('Failed to launch browser:', error);
throw error;
Expand Down Expand Up @@ -148,6 +151,7 @@ export class AIOBrowser {
async getActivePage(): Promise<Page> {
this.logger.info('Getting active page');
const pages = await this.browser.getBrowser().pages();
this.logger.info(`getActivePage pages length: ${pages.length}`);
try {
// First try to find a visible page without waiting
for (const page of pages) {
Expand Down Expand Up @@ -192,10 +196,14 @@ export class AIOBrowser {
continue;
}
}
this.logger.success('Active original page retrieved successfully');
this.logger.success('Active original page retrieved failed, fallback to active page');
return this.browser.getActivePage();
} catch (error) {
this.logger.error('Failed to get active page:', error);
if ((error as Error).message.includes('Protocol error: Connection closed')) {
this.logger.error('Connection closed, reconnecting...');
this.browser.launch(this.launchOptions);
}
throw error;
}
}
Expand Down
Loading
Loading