feat: browser_choose_file (#52)

Resolves https://github.com/microsoft/playwright-mcp/issues/31.

Prompt used for testing: 
```
upload test.txt to dropbox
```

This won't work for asynchronously opened filechoosers, but let's start
with the synchronous variant. I also tested with including the file
chooser open state in the snapshot, but that didn't give better results.

I also tested with OneDrive and WeTransfer, but somehow our ARIA
snapshots for those pages are missing some elements that are crucial to
locate the upload buttons.
This commit is contained in:
Simon Knott
2025-03-27 20:49:57 +01:00
committed by GitHub
parent f033213618
commit 5e200405e5
6 changed files with 152 additions and 10 deletions

View File

@@ -73,23 +73,36 @@ async function waitForCompletion<R>(page: playwright.Page, callback: () => Promi
export async function runAndWait(context: Context, status: string, callback: (page: playwright.Page) => Promise<any>, snapshot: boolean = false): Promise<ToolResult> {
const page = context.existingPage();
const dismissFileChooser = context.hasFileChooser();
await waitForCompletion(page, () => callback(page));
return snapshot ? captureAriaSnapshot(context, status) : {
if (dismissFileChooser)
context.clearFileChooser();
const result: ToolResult = snapshot ? await captureAriaSnapshot(context, status) : {
content: [{ type: 'text', text: status }],
};
return result;
}
export async function captureAriaSnapshot(context: Context, status: string = ''): Promise<ToolResult> {
const page = context.existingPage();
const lines = [];
if (status)
lines.push(`${status}`);
lines.push(
'',
`- Page URL: ${page.url()}`,
`- Page Title: ${await page.title()}`
);
if (context.hasFileChooser())
lines.push(`- There is a file chooser visible that requires browser_choose_file to be called`);
lines.push(
`- Page Snapshot`,
'```yaml',
await context.allFramesSnapshot(),
'```',
''
);
return {
content: [{ type: 'text', text: `${status ? `${status}\n` : ''}
- Page URL: ${page.url()}
- Page Title: ${await page.title()}
- Page Snapshot
\`\`\`yaml
${await context.allFramesSnapshot()}
\`\`\`
`
}],
content: [{ type: 'text', text: lines.join('\n') }],
};
}