figmar/BrowserSkill
1
0
Fork 0
mirror of https://github.com/Tencent/BrowserSkill.git synced 2026-09-10 13:01:23 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix(screenshot): fall back to CDP capture when captureVisibleTab fails (#83)

Browse Source 
chrome.tabs.captureVisibleTab reads back the window surface, which fails
outright on some Windows/Chrome combinations (Chromium
FAILURE_REASON_READBACK_FAILED: "Failed to capture tab: image readback
failed"), leaving full-tab screenshots without any working path.

Capture the full tab via CDP Page.captureScreenshot (fromSurface) when
the visible-tab capture rejects; the CDP path goes through the
renderer BeginFrame pipeline instead of the surface readback. When both
paths fail, return cdp_failed with data.reason=screenshot_capture_failed
carrying both underlying messages, and render a CLI hint that points at
the browser-side readback cause.

Refs #71
This commit is contained in:
BB-fat
2026-08-12 22:10:09 +08:00
committed by GitHub
parent 78c2719718
commit f81a1e7796
4 changed files with 157 additions and 23 deletions
Download Patch File Download Diff File Expand all files Collapse all files
apps/extension/src/tools/__tests__/observation.test.ts
+65
View File
@@ -196,6 +196,71 @@ describe("handleScreenshot", () => {
expect(res).toMatchObject({ code: "cdp_failed", message: /captureVisibleTab refused/ });
});
it("falls back to CDP Page.captureScreenshot when captureVisibleTab fails", async () => {
const sm = new SessionManager({ agentWindow: fakeAgentWindow([100]) });
await sm.start("aa11");
const capture = vi.fn(async () => {
throw new Error("Failed to capture tab: image readback failed");
});
const { cdp, sent } = makeFakeCdp({
"Page.captureScreenshot": () => ({ data: TINY_PNG }),
});
const res = await handleScreenshot(
sm,
{ session_id: "aa11" },
makeScreenshotDeps({ cdp, captureVisibleTab: capture }),
);
if ("code" in res) throw new Error(`unexpected error: ${JSON.stringify(res)}`);
expect(res.tab_id).toBe(7);
expect(res.image_base64).toBe(TINY_PNG);
expect(res.format).toBe("png");
expect(capture).toHaveBeenCalledTimes(1);
const fallbackCall = sent.find((c) => c.method === "Page.captureScreenshot");
expect(fallbackCall?.params).toEqual({ format: "png", fromSurface: true });
});
it("reports screenshot_capture_failed when both capture paths fail", async () => {
const sm = new SessionManager({ agentWindow: fakeAgentWindow([100]) });
await sm.start("aa11");
const capture = vi.fn(async () => {
throw new Error("Failed to capture tab: image readback failed");
});
const { cdp } = makeFakeCdp({
"Page.captureScreenshot": () => {
throw new Error("debugger detached");
},
});
const res = await handleScreenshot(
sm,
{ session_id: "aa11" },
makeScreenshotDeps({ cdp, captureVisibleTab: capture }),
);
expect(res).toMatchObject({
code: "cdp_failed",
data: { reason: "screenshot_capture_failed" },
});
const message = (res as { message?: string }).message ?? "";
expect(message).toContain("image readback failed");
expect(message).toContain("debugger detached");
});
it("falls back to CDP when captureVisibleTab returns and CDP yields data", async () => {
// Sanity: a successful primary path never touches CDP even when a
// runner is available.
const sm = new SessionManager({ agentWindow: fakeAgentWindow([100]) });
await sm.start("aa11");
const capture = vi.fn(async () => `data:image/png;base64,${TINY_PNG}`);
const { cdp, sent } = makeFakeCdp({});
const res = await handleScreenshot(
sm,
{ session_id: "aa11" },
makeScreenshotDeps({ cdp, captureVisibleTab: capture }),
);
if ("code" in res) throw new Error(`unexpected error: ${JSON.stringify(res)}`);
expect(res.image_base64).toBe(TINY_PNG);
expect(sent.find((c) => c.method === "Page.captureScreenshot")).toBeUndefined();
});
it("captures a clipped PNG when ref is given", async () => {
const sm = new SessionManager({ agentWindow: fakeAgentWindow([100]) });
const ctx = await sm.start("aa11");