@playwright/mcp 0.0.13 → 0.0.15

package/lib/tools/screen.js

@@ -14,133 +14,197 @@
  * See the License for the specific language governing permissions and
  * limitations under the License.
  */
+var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    var desc = Object.getOwnPropertyDescriptor(m, k);
+    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
+      desc = { enumerable: true, get: function() { return m[k]; } };
+    }
+    Object.defineProperty(o, k2, desc);
+}) : (function(o, m, k, k2) {
+    if (k2 === undefined) k2 = k;
+    o[k2] = m[k];
+}));
+var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
+    Object.defineProperty(o, "default", { enumerable: true, value: v });
+}) : function(o, v) {
+    o["default"] = v;
+});
+var __importStar = (this && this.__importStar) || (function () {
+    var ownKeys = function(o) {
+        ownKeys = Object.getOwnPropertyNames || function (o) {
+            var ar = [];
+            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
+            return ar;
+        };
+        return ownKeys(o);
+    };
+    return function (mod) {
+        if (mod && mod.__esModule) return mod;
+        var result = {};
+        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
+        __setModuleDefault(result, mod);
+        return result;
+    };
+})();
 Object.defineProperty(exports, "__esModule", { value: true });
 const zod_1 = require("zod");
-const zod_to_json_schema_1 = require("zod-to-json-schema");
-const screenshot = {
+const tool_1 = require("./tool");
+const javascript = __importStar(require("../javascript"));
+const elementSchema = zod_1.z.object({
+    element: zod_1.z.string().describe('Human-readable element description used to obtain permission to interact with the element'),
+});
+const screenshot = (0, tool_1.defineTool)({
     capability: 'core',
     schema: {
         name: 'browser_screen_capture',
         description: 'Take a screenshot of the current page',
-        inputSchema: (0, zod_to_json_schema_1.zodToJsonSchema)(zod_1.z.object({})),
+        inputSchema: zod_1.z.object({}),
     },
     handle: async (context) => {
         const tab = await context.ensureTab();
-        const screenshot = await tab.page.screenshot({ type: 'jpeg', quality: 50, scale: 'css' });
+        const options = { type: 'jpeg', quality: 50, scale: 'css' };
+        const code = [
+            `// Take a screenshot of the current page`,
+            `await page.screenshot(${javascript.formatObject(options)});`,
+        ];
+        const action = () => tab.page.screenshot(options).then(buffer => {
+            return {
+                content: [{ type: 'image', data: buffer.toString('base64'), mimeType: 'image/jpeg' }],
+            };
+        });
         return {
-            content: [{ type: 'image', data: screenshot.toString('base64'), mimeType: 'image/jpeg' }],
+            code,
+            action,
+            captureSnapshot: false,
+            waitForNetwork: false
         };
     },
-};
-const elementSchema = zod_1.z.object({
-    element: zod_1.z.string().describe('Human-readable element description used to obtain permission to interact with the element'),
-});
-const moveMouseSchema = elementSchema.extend({
-    x: zod_1.z.number().describe('X coordinate'),
-    y: zod_1.z.number().describe('Y coordinate'),
 });
-const moveMouse = {
+const moveMouse = (0, tool_1.defineTool)({
     capability: 'core',
     schema: {
         name: 'browser_screen_move_mouse',
         description: 'Move mouse to a given position',
-        inputSchema: (0, zod_to_json_schema_1.zodToJsonSchema)(moveMouseSchema),
+        inputSchema: elementSchema.extend({
+            x: zod_1.z.number().describe('X coordinate'),
+            y: zod_1.z.number().describe('Y coordinate'),
+        }),
     },
     handle: async (context, params) => {
-        const validatedParams = moveMouseSchema.parse(params);
-        const tab = context.currentTab();
-        await tab.page.mouse.move(validatedParams.x, validatedParams.y);
+        const tab = context.currentTabOrDie();
+        const code = [
+            `// Move mouse to (${params.x}, ${params.y})`,
+            `await page.mouse.move(${params.x}, ${params.y});`,
+        ];
+        const action = () => tab.page.mouse.move(params.x, params.y);
         return {
-            content: [{ type: 'text', text: `Moved mouse to (${validatedParams.x}, ${validatedParams.y})` }],
+            code,
+            action,
+            captureSnapshot: false,
+            waitForNetwork: false
         };
     },
-};
-const clickSchema = elementSchema.extend({
-    x: zod_1.z.number().describe('X coordinate'),
-    y: zod_1.z.number().describe('Y coordinate'),
 });
-const click = {
+const click = (0, tool_1.defineTool)({
     capability: 'core',
     schema: {
         name: 'browser_screen_click',
         description: 'Click left mouse button',
-        inputSchema: (0, zod_to_json_schema_1.zodToJsonSchema)(clickSchema),
+        inputSchema: elementSchema.extend({
+            x: zod_1.z.number().describe('X coordinate'),
+            y: zod_1.z.number().describe('Y coordinate'),
+        }),
     },
     handle: async (context, params) => {
-        return await context.currentTab().runAndWait(async (tab) => {
-            const validatedParams = clickSchema.parse(params);
-            const code = [
-                `// Click mouse at coordinates (${validatedParams.x}, ${validatedParams.y})`,
-                `await page.mouse.move(${validatedParams.x}, ${validatedParams.y});`,
-                `await page.mouse.down();`,
-                `await page.mouse.up();`,
-            ];
-            await tab.page.mouse.move(validatedParams.x, validatedParams.y);
+        const tab = context.currentTabOrDie();
+        const code = [
+            `// Click mouse at coordinates (${params.x}, ${params.y})`,
+            `await page.mouse.move(${params.x}, ${params.y});`,
+            `await page.mouse.down();`,
+            `await page.mouse.up();`,
+        ];
+        const action = async () => {
+            await tab.page.mouse.move(params.x, params.y);
             await tab.page.mouse.down();
             await tab.page.mouse.up();
-            return { code };
-        });
+        };
+        return {
+            code,
+            action,
+            captureSnapshot: false,
+            waitForNetwork: true,
+        };
     },
-};
-const dragSchema = elementSchema.extend({
-    startX: zod_1.z.number().describe('Start X coordinate'),
-    startY: zod_1.z.number().describe('Start Y coordinate'),
-    endX: zod_1.z.number().describe('End X coordinate'),
-    endY: zod_1.z.number().describe('End Y coordinate'),
 });
-const drag = {
+const drag = (0, tool_1.defineTool)({
     capability: 'core',
     schema: {
         name: 'browser_screen_drag',
         description: 'Drag left mouse button',
-        inputSchema: (0, zod_to_json_schema_1.zodToJsonSchema)(dragSchema),
+        inputSchema: elementSchema.extend({
+            startX: zod_1.z.number().describe('Start X coordinate'),
+            startY: zod_1.z.number().describe('Start Y coordinate'),
+            endX: zod_1.z.number().describe('End X coordinate'),
+            endY: zod_1.z.number().describe('End Y coordinate'),
+        }),
     },
     handle: async (context, params) => {
-        const validatedParams = dragSchema.parse(params);
-        return await context.currentTab().runAndWait(async (tab) => {
-            await tab.page.mouse.move(validatedParams.startX, validatedParams.startY);
+        const tab = context.currentTabOrDie();
+        const code = [
+            `// Drag mouse from (${params.startX}, ${params.startY}) to (${params.endX}, ${params.endY})`,
+            `await page.mouse.move(${params.startX}, ${params.startY});`,
+            `await page.mouse.down();`,
+            `await page.mouse.move(${params.endX}, ${params.endY});`,
+            `await page.mouse.up();`,
+        ];
+        const action = async () => {
+            await tab.page.mouse.move(params.startX, params.startY);
             await tab.page.mouse.down();
-            await tab.page.mouse.move(validatedParams.endX, validatedParams.endY);
+            await tab.page.mouse.move(params.endX, params.endY);
             await tab.page.mouse.up();
-            const code = [
-                `// Drag mouse from (${validatedParams.startX}, ${validatedParams.startY}) to (${validatedParams.endX}, ${validatedParams.endY})`,
-                `await page.mouse.move(${validatedParams.startX}, ${validatedParams.startY});`,
-                `await page.mouse.down();`,
-                `await page.mouse.move(${validatedParams.endX}, ${validatedParams.endY});`,
-                `await page.mouse.up();`,
-            ];
-            return { code };
-        });
+        };
+        return {
+            code,
+            action,
+            captureSnapshot: false,
+            waitForNetwork: true,
+        };
     },
-};
-const typeSchema = zod_1.z.object({
-    text: zod_1.z.string().describe('Text to type into the element'),
-    submit: zod_1.z.boolean().optional().describe('Whether to submit entered text (press Enter after)'),
 });
-const type = {
+const type = (0, tool_1.defineTool)({
     capability: 'core',
     schema: {
         name: 'browser_screen_type',
         description: 'Type text',
-        inputSchema: (0, zod_to_json_schema_1.zodToJsonSchema)(typeSchema),
+        inputSchema: zod_1.z.object({
+            text: zod_1.z.string().describe('Text to type into the element'),
+            submit: zod_1.z.boolean().optional().describe('Whether to submit entered text (press Enter after)'),
+        }),
     },
     handle: async (context, params) => {
-        const validatedParams = typeSchema.parse(params);
-        return await context.currentTab().runAndWait(async (tab) => {
-            const code = [
-                `// Type ${validatedParams.text}`,
-                `await page.keyboard.type('${validatedParams.text}');`,
-            ];
-            await tab.page.keyboard.type(validatedParams.text);
-            if (validatedParams.submit) {
-                code.push(`// Submit text`);
-                code.push(`await page.keyboard.press('Enter');`);
+        const tab = context.currentTabOrDie();
+        const code = [
+            `// Type ${params.text}`,
+            `await page.keyboard.type('${params.text}');`,
+        ];
+        const action = async () => {
+            await tab.page.keyboard.type(params.text);
+            if (params.submit)
                 await tab.page.keyboard.press('Enter');
-            }
-            return { code };
-        });
+        };
+        if (params.submit) {
+            code.push(`// Submit text`);
+            code.push(`await page.keyboard.press('Enter');`);
+        }
+        return {
+            code,
+            action,
+            captureSnapshot: false,
+            waitForNetwork: true,
+        };
     },
-};
+});
 exports.default = [
     screenshot,
     moveMouse,

package/lib/tools/snapshot.js

@@ -51,175 +51,220 @@ var __importDefault = (this && this.__importDefault) || function (mod) {
     return (mod && mod.__esModule) ? mod : { "default": mod };
 };
 Object.defineProperty(exports, "__esModule", { value: true });
+const path_1 = __importDefault(require("path"));
+const os_1 = __importDefault(require("os"));
 const zod_1 = require("zod");
-const zod_to_json_schema_1 = __importDefault(require("zod-to-json-schema"));
+const utils_1 = require("./utils");
 const context_1 = require("../context");
 const javascript = __importStar(require("../javascript"));
-const snapshot = {
+const tool_1 = require("./tool");
+const snapshot = (0, tool_1.defineTool)({
     capability: 'core',
     schema: {
         name: 'browser_snapshot',
         description: 'Capture accessibility snapshot of the current page, this is better than screenshot',
-        inputSchema: (0, zod_to_json_schema_1.default)(zod_1.z.object({})),
+        inputSchema: zod_1.z.object({}),
     },
     handle: async (context) => {
-        const tab = await context.ensureTab();
-        return await tab.run(async () => {
-            const code = [`// <internal code to capture accessibility snapshot>`];
-            return { code };
-        }, { captureSnapshot: true });
+        await context.ensureTab();
+        return {
+            code: [`// <internal code to capture accessibility snapshot>`],
+            captureSnapshot: true,
+            waitForNetwork: false,
+        };
     },
-};
+});
 const elementSchema = zod_1.z.object({
     element: zod_1.z.string().describe('Human-readable element description used to obtain permission to interact with the element'),
     ref: zod_1.z.string().describe('Exact target element reference from the page snapshot'),
 });
-const click = {
+const click = (0, tool_1.defineTool)({
     capability: 'core',
     schema: {
         name: 'browser_click',
         description: 'Perform click on a web page',
-        inputSchema: (0, zod_to_json_schema_1.default)(elementSchema),
+        inputSchema: elementSchema,
     },
     handle: async (context, params) => {
-        const validatedParams = elementSchema.parse(params);
-        return await context.currentTab().runAndWaitWithSnapshot(async (snapshot) => {
-            const locator = snapshot.refLocator(validatedParams.ref);
-            const code = [
-                `// Click ${validatedParams.element}`,
-                `await page.${await (0, context_1.generateLocator)(locator)}.click();`
-            ];
-            await locator.click();
-            return { code };
-        });
+        const tab = context.currentTabOrDie();
+        const locator = tab.snapshotOrDie().refLocator(params.ref);
+        const code = [
+            `// Click ${params.element}`,
+            `await page.${await (0, context_1.generateLocator)(locator)}.click();`
+        ];
+        return {
+            code,
+            action: () => locator.click(),
+            captureSnapshot: true,
+            waitForNetwork: true,
+        };
     },
-};
-const dragSchema = zod_1.z.object({
-    startElement: zod_1.z.string().describe('Human-readable source element description used to obtain the permission to interact with the element'),
-    startRef: zod_1.z.string().describe('Exact source element reference from the page snapshot'),
-    endElement: zod_1.z.string().describe('Human-readable target element description used to obtain the permission to interact with the element'),
-    endRef: zod_1.z.string().describe('Exact target element reference from the page snapshot'),
 });
-const drag = {
+const drag = (0, tool_1.defineTool)({
     capability: 'core',
     schema: {
         name: 'browser_drag',
         description: 'Perform drag and drop between two elements',
-        inputSchema: (0, zod_to_json_schema_1.default)(dragSchema),
+        inputSchema: zod_1.z.object({
+            startElement: zod_1.z.string().describe('Human-readable source element description used to obtain the permission to interact with the element'),
+            startRef: zod_1.z.string().describe('Exact source element reference from the page snapshot'),
+            endElement: zod_1.z.string().describe('Human-readable target element description used to obtain the permission to interact with the element'),
+            endRef: zod_1.z.string().describe('Exact target element reference from the page snapshot'),
+        }),
     },
     handle: async (context, params) => {
-        const validatedParams = dragSchema.parse(params);
-        return await context.currentTab().runAndWaitWithSnapshot(async (snapshot) => {
-            const startLocator = snapshot.refLocator(validatedParams.startRef);
-            const endLocator = snapshot.refLocator(validatedParams.endRef);
-            const code = [
-                `// Drag ${validatedParams.startElement} to ${validatedParams.endElement}`,
-                `await page.${await (0, context_1.generateLocator)(startLocator)}.dragTo(page.${await (0, context_1.generateLocator)(endLocator)});`
-            ];
-            await startLocator.dragTo(endLocator);
-            return { code };
-        });
+        const snapshot = context.currentTabOrDie().snapshotOrDie();
+        const startLocator = snapshot.refLocator(params.startRef);
+        const endLocator = snapshot.refLocator(params.endRef);
+        const code = [
+            `// Drag ${params.startElement} to ${params.endElement}`,
+            `await page.${await (0, context_1.generateLocator)(startLocator)}.dragTo(page.${await (0, context_1.generateLocator)(endLocator)});`
+        ];
+        return {
+            code,
+            action: () => startLocator.dragTo(endLocator),
+            captureSnapshot: true,
+            waitForNetwork: true,
+        };
     },
-};
-const hover = {
+});
+const hover = (0, tool_1.defineTool)({
     capability: 'core',
     schema: {
         name: 'browser_hover',
         description: 'Hover over element on page',
-        inputSchema: (0, zod_to_json_schema_1.default)(elementSchema),
+        inputSchema: elementSchema,
     },
     handle: async (context, params) => {
-        const validatedParams = elementSchema.parse(params);
-        return await context.currentTab().runAndWaitWithSnapshot(async (snapshot) => {
-            const locator = snapshot.refLocator(validatedParams.ref);
-            const code = [
-                `// Hover over ${validatedParams.element}`,
-                `await page.${await (0, context_1.generateLocator)(locator)}.hover();`
-            ];
-            await locator.hover();
-            return { code };
-        });
+        const snapshot = context.currentTabOrDie().snapshotOrDie();
+        const locator = snapshot.refLocator(params.ref);
+        const code = [
+            `// Hover over ${params.element}`,
+            `await page.${await (0, context_1.generateLocator)(locator)}.hover();`
+        ];
+        return {
+            code,
+            action: () => locator.hover(),
+            captureSnapshot: true,
+            waitForNetwork: true,
+        };
     },
-};
+});
 const typeSchema = elementSchema.extend({
     text: zod_1.z.string().describe('Text to type into the element'),
     submit: zod_1.z.boolean().optional().describe('Whether to submit entered text (press Enter after)'),
     slowly: zod_1.z.boolean().optional().describe('Whether to type one character at a time. Useful for triggering key handlers in the page. By default entire text is filled in at once.'),
 });
-const type = {
+const type = (0, tool_1.defineTool)({
     capability: 'core',
     schema: {
         name: 'browser_type',
         description: 'Type text into editable element',
-        inputSchema: (0, zod_to_json_schema_1.default)(typeSchema),
+        inputSchema: typeSchema,
     },
     handle: async (context, params) => {
-        const validatedParams = typeSchema.parse(params);
-        return await context.currentTab().runAndWaitWithSnapshot(async (snapshot) => {
-            const locator = snapshot.refLocator(validatedParams.ref);
-            const code = [];
-            if (validatedParams.slowly) {
-                code.push(`// Press "${validatedParams.text}" sequentially into "${validatedParams.element}"`);
-                code.push(`await page.${await (0, context_1.generateLocator)(locator)}.pressSequentially(${javascript.quote(validatedParams.text)});`);
-                await locator.pressSequentially(validatedParams.text);
-            }
-            else {
-                code.push(`// Fill "${validatedParams.text}" into "${validatedParams.element}"`);
-                code.push(`await page.${await (0, context_1.generateLocator)(locator)}.fill(${javascript.quote(validatedParams.text)});`);
-                await locator.fill(validatedParams.text);
-            }
-            if (validatedParams.submit) {
-                code.push(`// Submit text`);
-                code.push(`await page.${await (0, context_1.generateLocator)(locator)}.press('Enter');`);
-                await locator.press('Enter');
-            }
-            return { code };
-        });
+        const snapshot = context.currentTabOrDie().snapshotOrDie();
+        const locator = snapshot.refLocator(params.ref);
+        const code = [];
+        const steps = [];
+        if (params.slowly) {
+            code.push(`// Press "${params.text}" sequentially into "${params.element}"`);
+            code.push(`await page.${await (0, context_1.generateLocator)(locator)}.pressSequentially(${javascript.quote(params.text)});`);
+            steps.push(() => locator.pressSequentially(params.text));
+        }
+        else {
+            code.push(`// Fill "${params.text}" into "${params.element}"`);
+            code.push(`await page.${await (0, context_1.generateLocator)(locator)}.fill(${javascript.quote(params.text)});`);
+            steps.push(() => locator.fill(params.text));
+        }
+        if (params.submit) {
+            code.push(`// Submit text`);
+            code.push(`await page.${await (0, context_1.generateLocator)(locator)}.press('Enter');`);
+            steps.push(() => locator.press('Enter'));
+        }
+        return {
+            code,
+            action: () => steps.reduce((acc, step) => acc.then(step), Promise.resolve()),
+            captureSnapshot: true,
+            waitForNetwork: true,
+        };
     },
-};
+});
 const selectOptionSchema = elementSchema.extend({
     values: zod_1.z.array(zod_1.z.string()).describe('Array of values to select in the dropdown. This can be a single value or multiple values.'),
 });
-const selectOption = {
+const selectOption = (0, tool_1.defineTool)({
     capability: 'core',
     schema: {
         name: 'browser_select_option',
         description: 'Select an option in a dropdown',
-        inputSchema: (0, zod_to_json_schema_1.default)(selectOptionSchema),
+        inputSchema: selectOptionSchema,
     },
     handle: async (context, params) => {
-        const validatedParams = selectOptionSchema.parse(params);
-        return await context.currentTab().runAndWaitWithSnapshot(async (snapshot) => {
-            const locator = snapshot.refLocator(validatedParams.ref);
-            const code = [
-                `// Select options [${validatedParams.values.join(', ')}] in ${validatedParams.element}`,
-                `await page.${await (0, context_1.generateLocator)(locator)}.selectOption(${javascript.formatObject(validatedParams.values)});`
-            ];
-            await locator.selectOption(validatedParams.values);
-            return { code };
-        });
+        const snapshot = context.currentTabOrDie().snapshotOrDie();
+        const locator = snapshot.refLocator(params.ref);
+        const code = [
+            `// Select options [${params.values.join(', ')}] in ${params.element}`,
+            `await page.${await (0, context_1.generateLocator)(locator)}.selectOption(${javascript.formatObject(params.values)});`
+        ];
+        return {
+            code,
+            action: () => locator.selectOption(params.values).then(() => { }),
+            captureSnapshot: true,
+            waitForNetwork: true,
+        };
     },
-};
+});
 const screenshotSchema = zod_1.z.object({
     raw: zod_1.z.boolean().optional().describe('Whether to return without compression (in PNG format). Default is false, which returns a JPEG image.'),
+    element: zod_1.z.string().optional().describe('Human-readable element description used to obtain permission to screenshot the element. If not provided, the screenshot will be taken of viewport. If element is provided, ref must be provided too.'),
+    ref: zod_1.z.string().optional().describe('Exact target element reference from the page snapshot. If not provided, the screenshot will be taken of viewport. If ref is provided, element must be provided too.'),
+}).refine(data => {
+    return !!data.element === !!data.ref;
+}, {
+    message: 'Both element and ref must be provided or neither.',
+    path: ['ref', 'element']
 });
-const screenshot = {
+const screenshot = (0, tool_1.defineTool)({
     capability: 'core',
     schema: {
         name: 'browser_take_screenshot',
         description: `Take a screenshot of the current page. You can't perform actions based on the screenshot, use browser_snapshot for actions.`,
-        inputSchema: (0, zod_to_json_schema_1.default)(screenshotSchema),
+        inputSchema: screenshotSchema,
     },
     handle: async (context, params) => {
-        const validatedParams = screenshotSchema.parse(params);
-        const tab = context.currentTab();
-        const options = validatedParams.raw ? { type: 'png', scale: 'css' } : { type: 'jpeg', quality: 50, scale: 'css' };
-        const screenshot = await tab.page.screenshot(options);
+        const tab = context.currentTabOrDie();
+        const snapshot = tab.snapshotOrDie();
+        const fileType = params.raw ? 'png' : 'jpeg';
+        const fileName = path_1.default.join(os_1.default.tmpdir(), (0, utils_1.sanitizeForFilePath)(`page-${new Date().toISOString()}`)) + `.${fileType}`;
+        const options = { type: fileType, quality: fileType === 'png' ? undefined : 50, scale: 'css', path: fileName };
+        const isElementScreenshot = params.element && params.ref;
+        const code = [
+            `// Screenshot ${isElementScreenshot ? params.element : 'viewport'} and save it as ${fileName}`,
+        ];
+        const locator = params.ref ? snapshot.refLocator(params.ref) : null;
+        if (locator)
+            code.push(`await page.${await (0, context_1.generateLocator)(locator)}.screenshot(${javascript.formatObject(options)});`);
+        else
+            code.push(`await page.screenshot(${javascript.formatObject(options)});`);
+        const action = async () => {
+            const screenshot = locator ? await locator.screenshot(options) : await tab.page.screenshot(options);
+            return {
+                content: [{
+                        type: 'image',
+                        data: screenshot.toString('base64'),
+                        mimeType: fileType === 'png' ? 'image/png' : 'image/jpeg',
+                    }]
+            };
+        };
         return {
-            content: [{ type: 'image', data: screenshot.toString('base64'), mimeType: validatedParams.raw ? 'image/png' : 'image/jpeg' }],
+            code,
+            action,
+            captureSnapshot: true,
+            waitForNetwork: false,
         };
-    },
-};
+    }
+});
 exports.default = [
     snapshot,
     click,