feat(browser): add MCP tool dispatch

This commit is contained in:
Peter Steinberger
2025-12-19 23:57:26 +00:00
parent 0ac7a93c28
commit fa54950d2e
19 changed files with 2991 additions and 1243 deletions

View File

@@ -0,0 +1,50 @@
import type express from "express";
import type { BrowserRouteContext } from "../server-context.js";
import { jsonError } from "./utils.js";
export function registerBrowserBasicRoutes(
app: express.Express,
ctx: BrowserRouteContext,
) {
app.get("/", async (_req, res) => {
let current: ReturnType<typeof ctx.state>;
try {
current = ctx.state();
} catch {
return jsonError(res, 503, "browser server not started");
}
const reachable = await ctx.isReachable(300);
res.json({
enabled: current.resolved.enabled,
controlUrl: current.resolved.controlUrl,
running: reachable,
pid: current.running?.pid ?? null,
cdpPort: current.cdpPort,
chosenBrowser: current.running?.exe.kind ?? null,
userDataDir: current.running?.userDataDir ?? null,
color: current.resolved.color,
headless: current.resolved.headless,
attachOnly: current.resolved.attachOnly,
});
});
app.post("/start", async (_req, res) => {
try {
await ctx.ensureBrowserAvailable();
res.json({ ok: true });
} catch (err) {
jsonError(res, 500, String(err));
}
});
app.post("/stop", async (_req, res) => {
try {
const result = await ctx.stopRunningBrowser();
res.json({ ok: true, stopped: result.stopped });
} catch (err) {
jsonError(res, 500, String(err));
}
});
}

View File

@@ -0,0 +1,17 @@
import type express from "express";
import type { BrowserRouteContext } from "../server-context.js";
import { registerBrowserBasicRoutes } from "./basic.js";
import { registerBrowserInspectRoutes } from "./inspect.js";
import { registerBrowserTabRoutes } from "./tabs.js";
import { registerBrowserToolRoutes } from "./tool.js";
export function registerBrowserRoutes(
app: express.Express,
ctx: BrowserRouteContext,
) {
registerBrowserBasicRoutes(app, ctx);
registerBrowserTabRoutes(app, ctx);
registerBrowserInspectRoutes(app, ctx);
registerBrowserToolRoutes(app, ctx);
}

View File

@@ -0,0 +1,307 @@
import path from "node:path";
import type express from "express";
import { ensureMediaDir, saveMediaBuffer } from "../../media/store.js";
import {
captureScreenshot,
captureScreenshotPng,
evaluateJavaScript,
getDomText,
querySelector,
snapshotAria,
snapshotDom,
} from "../cdp.js";
import {
snapshotAiViaPlaywright,
takeScreenshotViaPlaywright,
} from "../pw-ai.js";
import {
DEFAULT_BROWSER_SCREENSHOT_MAX_BYTES,
DEFAULT_BROWSER_SCREENSHOT_MAX_SIDE,
normalizeBrowserScreenshot,
} from "../screenshot.js";
import type { BrowserRouteContext } from "../server-context.js";
import { jsonError, toBoolean, toStringOrEmpty } from "./utils.js";
export function registerBrowserInspectRoutes(
app: express.Express,
ctx: BrowserRouteContext,
) {
app.get("/screenshot", async (req, res) => {
const targetId =
typeof req.query.targetId === "string" ? req.query.targetId.trim() : "";
const fullPage =
req.query.fullPage === "true" || req.query.fullPage === "1";
try {
const tab = await ctx.ensureTabAvailable(targetId || undefined);
let shot: Buffer<ArrayBufferLike> = Buffer.alloc(0);
let contentTypeHint: "image/jpeg" | "image/png" = "image/jpeg";
try {
shot = await captureScreenshot({
wsUrl: tab.wsUrl ?? "",
fullPage,
format: "jpeg",
quality: 85,
});
} catch {
contentTypeHint = "image/png";
shot = await captureScreenshotPng({
wsUrl: tab.wsUrl ?? "",
fullPage,
});
}
const normalized = await normalizeBrowserScreenshot(shot, {
maxSide: DEFAULT_BROWSER_SCREENSHOT_MAX_SIDE,
maxBytes: DEFAULT_BROWSER_SCREENSHOT_MAX_BYTES,
});
await ensureMediaDir();
const saved = await saveMediaBuffer(
normalized.buffer,
normalized.contentType ?? contentTypeHint,
"browser",
DEFAULT_BROWSER_SCREENSHOT_MAX_BYTES,
);
const filePath = path.resolve(saved.path);
res.json({
ok: true,
path: filePath,
targetId: tab.targetId,
url: tab.url,
});
} catch (err) {
const mapped = ctx.mapTabError(err);
if (mapped) return jsonError(res, mapped.status, mapped.message);
jsonError(res, 500, String(err));
}
});
app.post("/screenshot", async (req, res) => {
const body = req.body as Record<string, unknown>;
const targetId = toStringOrEmpty(body?.targetId);
const fullPage = toBoolean(body?.fullPage) ?? false;
const ref = toStringOrEmpty(body?.ref);
const element = toStringOrEmpty(body?.element);
const type = body?.type === "jpeg" ? "jpeg" : "png";
const filename = toStringOrEmpty(body?.filename);
try {
const tab = await ctx.ensureTabAvailable(targetId || undefined);
const snap = await takeScreenshotViaPlaywright({
cdpPort: ctx.state().cdpPort,
targetId: tab.targetId,
ref,
element,
fullPage,
type,
});
const buffer = snap.buffer;
const normalized = await normalizeBrowserScreenshot(buffer, {
maxSide: DEFAULT_BROWSER_SCREENSHOT_MAX_SIDE,
maxBytes: DEFAULT_BROWSER_SCREENSHOT_MAX_BYTES,
});
await ensureMediaDir();
const saved = await saveMediaBuffer(
normalized.buffer,
normalized.contentType ?? `image/${type}`,
"browser",
DEFAULT_BROWSER_SCREENSHOT_MAX_BYTES,
);
const filePath = path.resolve(saved.path);
res.json({
ok: true,
path: filePath,
targetId: tab.targetId,
url: tab.url,
filename: filename || undefined,
});
} catch (err) {
const mapped = ctx.mapTabError(err);
if (mapped) return jsonError(res, mapped.status, mapped.message);
jsonError(res, 500, String(err));
}
});
app.post("/eval", async (req, res) => {
const js = toStringOrEmpty((req.body as { js?: unknown })?.js);
const targetId = toStringOrEmpty(
(req.body as { targetId?: unknown })?.targetId,
);
const awaitPromise = Boolean((req.body as { await?: unknown })?.await);
if (!js) return jsonError(res, 400, "js is required");
try {
const tab = await ctx.ensureTabAvailable(targetId || undefined);
const evaluated = await evaluateJavaScript({
wsUrl: tab.wsUrl ?? "",
expression: js,
awaitPromise,
returnByValue: true,
});
if (evaluated.exceptionDetails) {
const msg =
evaluated.exceptionDetails.exception?.description ||
evaluated.exceptionDetails.text ||
"JavaScript evaluation failed";
return jsonError(res, 400, msg);
}
res.json({
ok: true,
targetId: tab.targetId,
url: tab.url,
result: evaluated.result,
});
} catch (err) {
const mapped = ctx.mapTabError(err);
if (mapped) return jsonError(res, mapped.status, mapped.message);
jsonError(res, 500, String(err));
}
});
app.get("/query", async (req, res) => {
const selector =
typeof req.query.selector === "string" ? req.query.selector.trim() : "";
const targetId =
typeof req.query.targetId === "string" ? req.query.targetId.trim() : "";
const limit =
typeof req.query.limit === "string" ? Number(req.query.limit) : undefined;
if (!selector) return jsonError(res, 400, "selector is required");
try {
const tab = await ctx.ensureTabAvailable(targetId || undefined);
const result = await querySelector({
wsUrl: tab.wsUrl ?? "",
selector,
limit,
});
res.json({ ok: true, targetId: tab.targetId, url: tab.url, ...result });
} catch (err) {
const mapped = ctx.mapTabError(err);
if (mapped) return jsonError(res, mapped.status, mapped.message);
jsonError(res, 500, String(err));
}
});
app.get("/dom", async (req, res) => {
const targetId =
typeof req.query.targetId === "string" ? req.query.targetId.trim() : "";
const format = req.query.format === "text" ? "text" : "html";
const selector =
typeof req.query.selector === "string" ? req.query.selector.trim() : "";
const maxChars =
typeof req.query.maxChars === "string"
? Number(req.query.maxChars)
: undefined;
try {
const tab = await ctx.ensureTabAvailable(targetId || undefined);
const result = await getDomText({
wsUrl: tab.wsUrl ?? "",
format,
maxChars,
selector: selector || undefined,
});
res.json({
ok: true,
targetId: tab.targetId,
url: tab.url,
format,
...result,
});
} catch (err) {
const mapped = ctx.mapTabError(err);
if (mapped) return jsonError(res, mapped.status, mapped.message);
jsonError(res, 500, String(err));
}
});
app.get("/snapshot", async (req, res) => {
const targetId =
typeof req.query.targetId === "string" ? req.query.targetId.trim() : "";
const format =
req.query.format === "domSnapshot"
? "domSnapshot"
: req.query.format === "ai"
? "ai"
: "aria";
const limit =
typeof req.query.limit === "string" ? Number(req.query.limit) : undefined;
try {
const tab = await ctx.ensureTabAvailable(targetId || undefined);
if (format === "ai") {
const snap = await snapshotAiViaPlaywright({
cdpPort: ctx.state().cdpPort,
targetId: tab.targetId,
});
return res.json({
ok: true,
format,
targetId: tab.targetId,
url: tab.url,
...snap,
});
}
if (format === "aria") {
const snap = await snapshotAria({
wsUrl: tab.wsUrl ?? "",
limit,
});
return res.json({
ok: true,
format,
targetId: tab.targetId,
url: tab.url,
...snap,
});
}
const snap = await snapshotDom({
wsUrl: tab.wsUrl ?? "",
limit,
});
return res.json({
ok: true,
format,
targetId: tab.targetId,
url: tab.url,
...snap,
});
} catch (err) {
const mapped = ctx.mapTabError(err);
if (mapped) return jsonError(res, mapped.status, mapped.message);
jsonError(res, 500, String(err));
}
});
app.post("/click", async (req, res) => {
const ref = toStringOrEmpty((req.body as { ref?: unknown })?.ref);
const targetId = toStringOrEmpty(
(req.body as { targetId?: unknown })?.targetId,
);
if (!ref) return jsonError(res, 400, "ref is required");
try {
const tab = await ctx.ensureTabAvailable(targetId || undefined);
await clickViaPlaywright({
cdpPort: ctx.state().cdpPort,
targetId: tab.targetId,
ref,
});
res.json({ ok: true, targetId: tab.targetId, url: tab.url });
} catch (err) {
const mapped = ctx.mapTabError(err);
if (mapped) return jsonError(res, mapped.status, mapped.message);
jsonError(res, 500, String(err));
}
});
}

108
src/browser/routes/tabs.ts Normal file
View File

@@ -0,0 +1,108 @@
import type express from "express";
import type { BrowserRouteContext } from "../server-context.js";
import { jsonError, toNumber, toStringOrEmpty } from "./utils.js";
export function registerBrowserTabRoutes(
app: express.Express,
ctx: BrowserRouteContext,
) {
app.get("/tabs", async (_req, res) => {
try {
const reachable = await ctx.isReachable(300);
if (!reachable)
return res.json({ running: false, tabs: [] as unknown[] });
const tabs = await ctx.listTabs();
res.json({ running: true, tabs });
} catch (err) {
jsonError(res, 500, String(err));
}
});
app.post("/tabs/open", async (req, res) => {
const url = toStringOrEmpty((req.body as { url?: unknown })?.url);
if (!url) return jsonError(res, 400, "url is required");
try {
await ctx.ensureBrowserAvailable();
const tab = await ctx.openTab(url);
res.json(tab);
} catch (err) {
jsonError(res, 500, String(err));
}
});
app.post("/tabs/focus", async (req, res) => {
const targetId = toStringOrEmpty(
(req.body as { targetId?: unknown })?.targetId,
);
if (!targetId) return jsonError(res, 400, "targetId is required");
try {
if (!(await ctx.isReachable(300)))
return jsonError(res, 409, "browser not running");
await ctx.focusTab(targetId);
res.json({ ok: true });
} catch (err) {
const mapped = ctx.mapTabError(err);
if (mapped) return jsonError(res, mapped.status, mapped.message);
jsonError(res, 500, String(err));
}
});
app.delete("/tabs/:targetId", async (req, res) => {
const targetId = toStringOrEmpty(req.params.targetId);
if (!targetId) return jsonError(res, 400, "targetId is required");
try {
if (!(await ctx.isReachable(300)))
return jsonError(res, 409, "browser not running");
await ctx.closeTab(targetId);
res.json({ ok: true });
} catch (err) {
const mapped = ctx.mapTabError(err);
if (mapped) return jsonError(res, mapped.status, mapped.message);
jsonError(res, 500, String(err));
}
});
app.post("/tabs/action", async (req, res) => {
const action = toStringOrEmpty((req.body as { action?: unknown })?.action);
const index = toNumber((req.body as { index?: unknown })?.index);
try {
if (action === "list") {
const reachable = await ctx.isReachable(300);
if (!reachable) return res.json({ ok: true, tabs: [] as unknown[] });
const tabs = await ctx.listTabs();
return res.json({ ok: true, tabs });
}
if (action === "new") {
await ctx.ensureBrowserAvailable();
const tab = await ctx.openTab("about:blank");
return res.json({ ok: true, tab });
}
if (action === "close") {
const tabs = await ctx.listTabs();
const target = typeof index === "number" ? tabs[index] : tabs.at(0);
if (!target) return jsonError(res, 404, "tab not found");
await ctx.closeTab(target.targetId);
return res.json({ ok: true, targetId: target.targetId });
}
if (action === "select") {
if (typeof index !== "number")
return jsonError(res, 400, "index is required");
const tabs = await ctx.listTabs();
const target = tabs[index];
if (!target) return jsonError(res, 404, "tab not found");
await ctx.focusTab(target.targetId);
return res.json({ ok: true, targetId: target.targetId });
}
return jsonError(res, 400, "unknown tab action");
} catch (err) {
const mapped = ctx.mapTabError(err);
if (mapped) return jsonError(res, mapped.status, mapped.message);
jsonError(res, 500, String(err));
}
});
}

View File

@@ -0,0 +1,432 @@
import path from "node:path";
import type express from "express";
import { ensureMediaDir, saveMediaBuffer } from "../../media/store.js";
import {
clickViaPlaywright,
closePageViaPlaywright,
dragViaPlaywright,
evaluateViaPlaywright,
fileUploadViaPlaywright,
fillFormViaPlaywright,
handleDialogViaPlaywright,
hoverViaPlaywright,
navigateBackViaPlaywright,
navigateViaPlaywright,
pressKeyViaPlaywright,
resizeViewportViaPlaywright,
runCodeViaPlaywright,
selectOptionViaPlaywright,
snapshotAiViaPlaywright,
takeScreenshotViaPlaywright,
typeViaPlaywright,
waitForViaPlaywright,
} from "../pw-ai.js";
import {
DEFAULT_BROWSER_SCREENSHOT_MAX_BYTES,
DEFAULT_BROWSER_SCREENSHOT_MAX_SIDE,
normalizeBrowserScreenshot,
} from "../screenshot.js";
import type { BrowserRouteContext } from "../server-context.js";
import {
jsonError,
toBoolean,
toNumber,
toStringArray,
toStringOrEmpty,
} from "./utils.js";
type ToolCoreParams = {
name: string;
args: Record<string, unknown>;
targetId: string;
cdpPort: number;
ctx: BrowserRouteContext;
res: express.Response;
};
export async function handleBrowserToolCore(
params: ToolCoreParams,
): Promise<boolean> {
const { name, args, targetId, cdpPort, ctx, res } = params;
const target = targetId || undefined;
switch (name) {
case "browser_close": {
const tab = await ctx.ensureTabAvailable(target);
await closePageViaPlaywright({ cdpPort, targetId: tab.targetId });
res.json({ ok: true, targetId: tab.targetId, url: tab.url });
return true;
}
case "browser_resize": {
const width = toNumber(args.width);
const height = toNumber(args.height);
if (!width || !height) {
jsonError(res, 400, "width and height are required");
return true;
}
const tab = await ctx.ensureTabAvailable(target);
await resizeViewportViaPlaywright({
cdpPort,
targetId: tab.targetId,
width,
height,
});
res.json({ ok: true, targetId: tab.targetId, url: tab.url });
return true;
}
case "browser_handle_dialog": {
const accept = toBoolean(args.accept);
if (accept === undefined) {
jsonError(res, 400, "accept is required");
return true;
}
const promptText = toStringOrEmpty(args.promptText) || undefined;
const tab = await ctx.ensureTabAvailable(target);
const result = await handleDialogViaPlaywright({
cdpPort,
targetId: tab.targetId,
accept,
promptText,
});
res.json({ ok: true, ...result });
return true;
}
case "browser_evaluate": {
const fn = toStringOrEmpty(args.function);
if (!fn) {
jsonError(res, 400, "function is required");
return true;
}
const ref = toStringOrEmpty(args.ref) || undefined;
const tab = await ctx.ensureTabAvailable(target);
const result = await evaluateViaPlaywright({
cdpPort,
targetId: tab.targetId,
fn,
ref,
});
res.json({ ok: true, result });
return true;
}
case "browser_file_upload": {
const paths = toStringArray(args.paths) ?? [];
const tab = await ctx.ensureTabAvailable(target);
await fileUploadViaPlaywright({
cdpPort,
targetId: tab.targetId,
paths: paths.length ? paths : undefined,
});
res.json({ ok: true, targetId: tab.targetId });
return true;
}
case "browser_fill_form": {
const fields = Array.isArray(args.fields)
? (args.fields as Array<Record<string, unknown>>)
: null;
if (!fields?.length) {
jsonError(res, 400, "fields are required");
return true;
}
const tab = await ctx.ensureTabAvailable(target);
await fillFormViaPlaywright({
cdpPort,
targetId: tab.targetId,
fields,
});
res.json({ ok: true, targetId: tab.targetId });
return true;
}
case "browser_install": {
res.json({
ok: true,
message:
"clawd browser uses system Chrome/Chromium; no Playwright install needed.",
});
return true;
}
case "browser_press_key": {
const key = toStringOrEmpty(args.key);
if (!key) {
jsonError(res, 400, "key is required");
return true;
}
const tab = await ctx.ensureTabAvailable(target);
await pressKeyViaPlaywright({
cdpPort,
targetId: tab.targetId,
key,
});
res.json({ ok: true, targetId: tab.targetId });
return true;
}
case "browser_type": {
const ref = toStringOrEmpty(args.ref);
const text = toStringOrEmpty(args.text);
if (!ref || !text) {
jsonError(res, 400, "ref and text are required");
return true;
}
const submit = toBoolean(args.submit) ?? false;
const slowly = toBoolean(args.slowly) ?? false;
const tab = await ctx.ensureTabAvailable(target);
await typeViaPlaywright({
cdpPort,
targetId: tab.targetId,
ref,
text,
submit,
slowly,
});
res.json({ ok: true, targetId: tab.targetId });
return true;
}
case "browser_navigate": {
const url = toStringOrEmpty(args.url);
if (!url) {
jsonError(res, 400, "url is required");
return true;
}
const tab = await ctx.ensureTabAvailable(target);
const result = await navigateViaPlaywright({
cdpPort,
targetId: tab.targetId,
url,
});
res.json({ ok: true, targetId: tab.targetId, ...result });
return true;
}
case "browser_navigate_back": {
const tab = await ctx.ensureTabAvailable(target);
const result = await navigateBackViaPlaywright({
cdpPort,
targetId: tab.targetId,
});
res.json({ ok: true, targetId: tab.targetId, ...result });
return true;
}
case "browser_run_code": {
const code = toStringOrEmpty(args.code);
if (!code) {
jsonError(res, 400, "code is required");
return true;
}
const tab = await ctx.ensureTabAvailable(target);
const result = await runCodeViaPlaywright({
cdpPort,
targetId: tab.targetId,
code,
});
res.json({ ok: true, result });
return true;
}
case "browser_take_screenshot": {
const type = args.type === "jpeg" ? "jpeg" : "png";
const ref = toStringOrEmpty(args.ref) || undefined;
const fullPage = toBoolean(args.fullPage) ?? false;
const element = toStringOrEmpty(args.element) || undefined;
const filename = toStringOrEmpty(args.filename) || undefined;
const tab = await ctx.ensureTabAvailable(target);
const snap = await takeScreenshotViaPlaywright({
cdpPort,
targetId: tab.targetId,
ref,
element,
fullPage,
type,
});
const normalized = await normalizeBrowserScreenshot(snap.buffer, {
maxSide: DEFAULT_BROWSER_SCREENSHOT_MAX_SIDE,
maxBytes: DEFAULT_BROWSER_SCREENSHOT_MAX_BYTES,
});
await ensureMediaDir();
const saved = await saveMediaBuffer(
normalized.buffer,
normalized.contentType ?? `image/${type}`,
"browser",
DEFAULT_BROWSER_SCREENSHOT_MAX_BYTES,
);
res.json({
ok: true,
path: path.resolve(saved.path),
filename,
targetId: tab.targetId,
url: tab.url,
});
return true;
}
case "browser_snapshot": {
const filename = toStringOrEmpty(args.filename) || undefined;
const tab = await ctx.ensureTabAvailable(target);
const snap = await snapshotAiViaPlaywright({
cdpPort,
targetId: tab.targetId,
});
if (filename) {
await ensureMediaDir();
const saved = await saveMediaBuffer(
Buffer.from(snap.snapshot, "utf8"),
"text/plain",
"browser",
);
res.json({
ok: true,
path: path.resolve(saved.path),
filename,
targetId: tab.targetId,
url: tab.url,
});
return true;
}
res.json({
ok: true,
snapshot: snap.snapshot,
targetId: tab.targetId,
url: tab.url,
});
return true;
}
case "browser_click": {
const ref = toStringOrEmpty(args.ref);
if (!ref) {
jsonError(res, 400, "ref is required");
return true;
}
const doubleClick = toBoolean(args.doubleClick) ?? false;
const button = toStringOrEmpty(args.button) || undefined;
const modifiers = Array.isArray(args.modifiers)
? (args.modifiers as string[])
: undefined;
const tab = await ctx.ensureTabAvailable(target);
await clickViaPlaywright({
cdpPort,
targetId: tab.targetId,
ref,
doubleClick,
button,
modifiers,
});
res.json({ ok: true, targetId: tab.targetId, url: tab.url });
return true;
}
case "browser_drag": {
const startRef = toStringOrEmpty(args.startRef);
const endRef = toStringOrEmpty(args.endRef);
if (!startRef || !endRef) {
jsonError(res, 400, "startRef and endRef are required");
return true;
}
const tab = await ctx.ensureTabAvailable(target);
await dragViaPlaywright({
cdpPort,
targetId: tab.targetId,
startRef,
endRef,
});
res.json({ ok: true, targetId: tab.targetId });
return true;
}
case "browser_hover": {
const ref = toStringOrEmpty(args.ref);
if (!ref) {
jsonError(res, 400, "ref is required");
return true;
}
const tab = await ctx.ensureTabAvailable(target);
await hoverViaPlaywright({
cdpPort,
targetId: tab.targetId,
ref,
});
res.json({ ok: true, targetId: tab.targetId });
return true;
}
case "browser_select_option": {
const ref = toStringOrEmpty(args.ref);
const values = toStringArray(args.values);
if (!ref || !values?.length) {
jsonError(res, 400, "ref and values are required");
return true;
}
const tab = await ctx.ensureTabAvailable(target);
await selectOptionViaPlaywright({
cdpPort,
targetId: tab.targetId,
ref,
values,
});
res.json({ ok: true, targetId: tab.targetId });
return true;
}
case "browser_tabs": {
const action = toStringOrEmpty(args.action);
const index = toNumber(args.index);
if (!action) {
jsonError(res, 400, "action is required");
return true;
}
if (action === "list") {
const reachable = await ctx.isReachable(300);
if (!reachable) {
res.json({ ok: true, tabs: [] });
return true;
}
const tabs = await ctx.listTabs();
res.json({ ok: true, tabs });
return true;
}
if (action === "new") {
await ctx.ensureBrowserAvailable();
const tab = await ctx.openTab("about:blank");
res.json({ ok: true, tab });
return true;
}
if (action === "close") {
const tabs = await ctx.listTabs();
const targetTab = typeof index === "number" ? tabs[index] : tabs.at(0);
if (!targetTab) {
jsonError(res, 404, "tab not found");
return true;
}
await ctx.closeTab(targetTab.targetId);
res.json({ ok: true, targetId: targetTab.targetId });
return true;
}
if (action === "select") {
if (typeof index !== "number") {
jsonError(res, 400, "index is required");
return true;
}
const tabs = await ctx.listTabs();
const targetTab = tabs[index];
if (!targetTab) {
jsonError(res, 404, "tab not found");
return true;
}
await ctx.focusTab(targetTab.targetId);
res.json({ ok: true, targetId: targetTab.targetId });
return true;
}
jsonError(res, 400, "unknown tab action");
return true;
}
case "browser_wait_for": {
const time = toNumber(args.time);
const text = toStringOrEmpty(args.text) || undefined;
const textGone = toStringOrEmpty(args.textGone) || undefined;
const tab = await ctx.ensureTabAvailable(target);
await waitForViaPlaywright({
cdpPort,
targetId: tab.targetId,
time,
text,
textGone,
});
res.json({ ok: true, targetId: tab.targetId });
return true;
}
default:
return false;
}
}

View File

@@ -0,0 +1,262 @@
import path from "node:path";
import type express from "express";
import { ensureMediaDir, saveMediaBuffer } from "../../media/store.js";
import {
generateLocatorForRef,
getConsoleMessagesViaPlaywright,
getNetworkRequestsViaPlaywright,
mouseClickViaPlaywright,
mouseDragViaPlaywright,
mouseMoveViaPlaywright,
pdfViaPlaywright,
startTracingViaPlaywright,
stopTracingViaPlaywright,
verifyElementVisibleViaPlaywright,
verifyListVisibleViaPlaywright,
verifyTextVisibleViaPlaywright,
verifyValueViaPlaywright,
} from "../pw-ai.js";
import type { BrowserRouteContext } from "../server-context.js";
import {
jsonError,
toBoolean,
toNumber,
toStringArray,
toStringOrEmpty,
} from "./utils.js";
type ToolExtraParams = {
name: string;
args: Record<string, unknown>;
targetId: string;
cdpPort: number;
ctx: BrowserRouteContext;
res: express.Response;
};
export async function handleBrowserToolExtra(
params: ToolExtraParams,
): Promise<boolean> {
const { name, args, targetId, cdpPort, ctx, res } = params;
const target = targetId || undefined;
switch (name) {
case "browser_console_messages": {
const level = toStringOrEmpty(args.level) || undefined;
const tab = await ctx.ensureTabAvailable(target);
const messages = await getConsoleMessagesViaPlaywright({
cdpPort,
targetId: tab.targetId,
level,
});
res.json({ ok: true, messages, targetId: tab.targetId });
return true;
}
case "browser_network_requests": {
const includeStatic = toBoolean(args.includeStatic) ?? false;
const tab = await ctx.ensureTabAvailable(target);
const requests = await getNetworkRequestsViaPlaywright({
cdpPort,
targetId: tab.targetId,
includeStatic,
});
res.json({ ok: true, requests, targetId: tab.targetId });
return true;
}
case "browser_pdf_save": {
const tab = await ctx.ensureTabAvailable(target);
const pdf = await pdfViaPlaywright({
cdpPort,
targetId: tab.targetId,
});
await ensureMediaDir();
const saved = await saveMediaBuffer(
pdf.buffer,
"application/pdf",
"browser",
pdf.buffer.byteLength,
);
res.json({
ok: true,
path: path.resolve(saved.path),
targetId: tab.targetId,
url: tab.url,
});
return true;
}
case "browser_start_tracing": {
const tab = await ctx.ensureTabAvailable(target);
await startTracingViaPlaywright({
cdpPort,
targetId: tab.targetId,
});
res.json({ ok: true });
return true;
}
case "browser_stop_tracing": {
const tab = await ctx.ensureTabAvailable(target);
const trace = await stopTracingViaPlaywright({
cdpPort,
targetId: tab.targetId,
});
await ensureMediaDir();
const saved = await saveMediaBuffer(
trace.buffer,
"application/zip",
"browser",
trace.buffer.byteLength,
);
res.json({
ok: true,
path: path.resolve(saved.path),
targetId: tab.targetId,
url: tab.url,
});
return true;
}
case "browser_verify_element_visible": {
const role = toStringOrEmpty(args.role);
const accessibleName = toStringOrEmpty(args.accessibleName);
if (!role || !accessibleName) {
jsonError(res, 400, "role and accessibleName are required");
return true;
}
const tab = await ctx.ensureTabAvailable(target);
await verifyElementVisibleViaPlaywright({
cdpPort,
targetId: tab.targetId,
role,
accessibleName,
});
res.json({ ok: true });
return true;
}
case "browser_verify_text_visible": {
const text = toStringOrEmpty(args.text);
if (!text) {
jsonError(res, 400, "text is required");
return true;
}
const tab = await ctx.ensureTabAvailable(target);
await verifyTextVisibleViaPlaywright({
cdpPort,
targetId: tab.targetId,
text,
});
res.json({ ok: true });
return true;
}
case "browser_verify_list_visible": {
const ref = toStringOrEmpty(args.ref);
const items = toStringArray(args.items);
if (!ref || !items?.length) {
jsonError(res, 400, "ref and items are required");
return true;
}
const tab = await ctx.ensureTabAvailable(target);
await verifyListVisibleViaPlaywright({
cdpPort,
targetId: tab.targetId,
ref,
items,
});
res.json({ ok: true });
return true;
}
case "browser_verify_value": {
const ref = toStringOrEmpty(args.ref);
const type = toStringOrEmpty(args.type);
const value = toStringOrEmpty(args.value);
if (!ref || !type) {
jsonError(res, 400, "ref and type are required");
return true;
}
const tab = await ctx.ensureTabAvailable(target);
await verifyValueViaPlaywright({
cdpPort,
targetId: tab.targetId,
ref,
type,
value,
});
res.json({ ok: true });
return true;
}
case "browser_mouse_move_xy": {
const x = toNumber(args.x);
const y = toNumber(args.y);
if (x === undefined || y === undefined) {
jsonError(res, 400, "x and y are required");
return true;
}
const tab = await ctx.ensureTabAvailable(target);
await mouseMoveViaPlaywright({
cdpPort,
targetId: tab.targetId,
x,
y,
});
res.json({ ok: true });
return true;
}
case "browser_mouse_click_xy": {
const x = toNumber(args.x);
const y = toNumber(args.y);
if (x === undefined || y === undefined) {
jsonError(res, 400, "x and y are required");
return true;
}
const button = toStringOrEmpty(args.button) || undefined;
const tab = await ctx.ensureTabAvailable(target);
await mouseClickViaPlaywright({
cdpPort,
targetId: tab.targetId,
x,
y,
button,
});
res.json({ ok: true });
return true;
}
case "browser_mouse_drag_xy": {
const startX = toNumber(args.startX);
const startY = toNumber(args.startY);
const endX = toNumber(args.endX);
const endY = toNumber(args.endY);
if (
startX === undefined ||
startY === undefined ||
endX === undefined ||
endY === undefined
) {
jsonError(res, 400, "startX, startY, endX, endY are required");
return true;
}
const tab = await ctx.ensureTabAvailable(target);
await mouseDragViaPlaywright({
cdpPort,
targetId: tab.targetId,
startX,
startY,
endX,
endY,
});
res.json({ ok: true });
return true;
}
case "browser_generate_locator": {
const ref = toStringOrEmpty(args.ref);
if (!ref) {
jsonError(res, 400, "ref is required");
return true;
}
const locator = generateLocatorForRef(ref);
res.json({ ok: true, locator });
return true;
}
default:
return false;
}
}

View File

@@ -0,0 +1,65 @@
import type express from "express";
import type { BrowserRouteContext } from "../server-context.js";
import { handleBrowserToolCore } from "./tool-core.js";
import { handleBrowserToolExtra } from "./tool-extra.js";
import { jsonError, toStringOrEmpty } from "./utils.js";
type ToolRequestBody = {
name?: unknown;
args?: unknown;
targetId?: unknown;
};
function toolArgs(value: unknown): Record<string, unknown> {
if (!value || typeof value !== "object" || Array.isArray(value)) return {};
return value as Record<string, unknown>;
}
export function registerBrowserToolRoutes(
app: express.Express,
ctx: BrowserRouteContext,
) {
app.post("/tool", async (req, res) => {
const body = req.body as ToolRequestBody;
const name = toStringOrEmpty(body?.name);
if (!name) return jsonError(res, 400, "name is required");
const args = toolArgs(body?.args);
const targetId = toStringOrEmpty(body?.targetId || args?.targetId);
try {
let cdpPort: number;
try {
cdpPort = ctx.state().cdpPort;
} catch {
return jsonError(res, 503, "browser server not started");
}
const handledCore = await handleBrowserToolCore({
name,
args,
targetId,
cdpPort,
ctx,
res,
});
if (handledCore) return;
const handledExtra = await handleBrowserToolExtra({
name,
args,
targetId,
cdpPort,
ctx,
res,
});
if (handledExtra) return;
return jsonError(res, 400, "unknown tool name");
} catch (err) {
const mapped = ctx.mapTabError(err);
if (mapped) return jsonError(res, mapped.status, mapped.message);
jsonError(res, 500, String(err));
}
});
}

View File

@@ -0,0 +1,38 @@
import type express from "express";
export function jsonError(
res: express.Response,
status: number,
message: string,
) {
res.status(status).json({ error: message });
}
export function toStringOrEmpty(value: unknown) {
return typeof value === "string" ? value.trim() : String(value ?? "").trim();
}
export function toNumber(value: unknown) {
if (typeof value === "number" && Number.isFinite(value)) return value;
if (typeof value === "string" && value.trim()) {
const parsed = Number(value);
return Number.isFinite(parsed) ? parsed : undefined;
}
return undefined;
}
export function toBoolean(value: unknown) {
if (typeof value === "boolean") return value;
if (typeof value === "string") {
const v = value.trim().toLowerCase();
if (v === "true" || v === "1" || v === "yes") return true;
if (v === "false" || v === "0" || v === "no") return false;
}
return undefined;
}
export function toStringArray(value: unknown): string[] | undefined {
if (!Array.isArray(value)) return undefined;
const strings = value.map((v) => toStringOrEmpty(v)).filter(Boolean);
return strings.length ? strings : undefined;
}