Browser tools: full accessibility snapshot, trusted input, dialogs, evaluate, and tab controls (#1881)

* feat(browser): replace flat snapshot with ARIA tree and page-side refs

* feat(browser): trusted CDP input with actionability waits

* feat(browser): auto-handle javascript dialogs

* feat(browser): add browser_evaluate

* feat(browser): add scroll, resize, and close_tab tools

* fix(browser): review fixes for snapshot/input/dialog seams

* fix(browser): address review findings on dialog capture, truncation, and keypress targets

* fix(harness): drop resize check that cannot run under parked compositor surfaces

* fix(browser): second review round — snapshot fidelity, key input, dialog scoping
This commit is contained in:
Mohamed Boudra
2026-07-04 01:38:07 +02:00
committed by GitHub
parent b28f291493
commit 9c220c2451
29 changed files with 5653 additions and 949 deletions
@@ -4,6 +4,7 @@ const RESIDENT_VIEWPORT_WIDTH = 1280;
const RESIDENT_VIEWPORT_HEIGHT = 800;
const residentWebviewsByBrowserId = new Map<string, HTMLElement>();
const residentWebviewSizesByBrowserId = new Map<string, { width: number; height: number }>();
interface BrowserWebviewElement extends HTMLElement {
src: string;
@@ -66,11 +67,24 @@ function findBrowserWebview(browserId: string, ownerDocument: Document): HTMLEle
return null;
}
function applyResidentWebviewStyle(webview: HTMLElement): void {
function dimensionsForBrowser(browserId: string | null): { width: number; height: number } {
if (!browserId) {
return { width: RESIDENT_VIEWPORT_WIDTH, height: RESIDENT_VIEWPORT_HEIGHT };
}
return (
residentWebviewSizesByBrowserId.get(browserId) ?? {
width: RESIDENT_VIEWPORT_WIDTH,
height: RESIDENT_VIEWPORT_HEIGHT,
}
);
}
function applyResidentWebviewStyle(webview: HTMLElement, browserId: string | null): void {
const dimensions = dimensionsForBrowser(browserId);
webview.style.display = "inline-flex";
webview.style.flex = "0 0 auto";
webview.style.width = `${RESIDENT_VIEWPORT_WIDTH}px`;
webview.style.height = `${RESIDENT_VIEWPORT_HEIGHT}px`;
webview.style.width = `${dimensions.width}px`;
webview.style.height = `${dimensions.height}px`;
webview.style.border = "0";
webview.style.background = "transparent";
webview.style.position = "absolute";
@@ -163,10 +177,33 @@ export function releaseResidentBrowserWebview(browserId: string, webview: HTMLEl
}
residentWebviewsByBrowserId.set(normalizedBrowserId, webview);
applyResidentWebviewStyle(webview);
applyResidentWebviewStyle(webview, normalizedBrowserId);
getResidentBrowserHost(ownerDocument).appendChild(webview);
}
export function resizeResidentBrowserWebview(input: {
browserId: string;
width: number;
height: number;
}): { width: number; height: number } | null {
const normalizedBrowserId = trimNonEmpty(input.browserId);
if (!normalizedBrowserId) {
return null;
}
const width = Math.max(1, Math.round(input.width));
const height = Math.max(1, Math.round(input.height));
residentWebviewSizesByBrowserId.set(normalizedBrowserId, { width, height });
const ownerDocument = readDocument();
const webview = ownerDocument ? findBrowserWebview(normalizedBrowserId, ownerDocument) : null;
if (webview) {
webview.style.width = `${width}px`;
webview.style.height = `${height}px`;
}
return { width, height };
}
export function removeResidentBrowserWebview(browserId: string): void {
const normalizedBrowserId = trimNonEmpty(browserId);
if (!normalizedBrowserId) {
@@ -175,6 +212,7 @@ export function removeResidentBrowserWebview(browserId: string): void {
const resident = residentWebviewsByBrowserId.get(normalizedBrowserId) ?? null;
residentWebviewsByBrowserId.delete(normalizedBrowserId);
residentWebviewSizesByBrowserId.delete(normalizedBrowserId);
resident?.remove();
}
@@ -183,5 +221,6 @@ export function clearResidentBrowserWebviewsForTests(): void {
webview.remove();
}
residentWebviewsByBrowserId.clear();
residentWebviewSizesByBrowserId.clear();
readDocument()?.getElementById(RESIDENT_BROWSER_HOST_ID)?.remove();
}