@mastra/agent-browser 0.3.0-alpha.0 → 0.3.0-alpha.1

package/CHANGELOG.md

@@ -1,5 +1,24 @@
 # @mastra/agent-browser
 
+## 0.3.0-alpha.1
+
+### Minor Changes
+
+- Add `waitUntil` support to `browser_click`, `browser_press`, and `browser_select`. When provided, the tool waits for the page to reach the given load state (`load`, `domcontentloaded`, or `networkidle`) after the action completes, preventing the next `browser_snapshot` from capturing stale DOM when the interaction triggers navigation. The parameter is optional and behaviour is unchanged when omitted. ([#17426](https://github.com/mastra-ai/mastra/pull/17426))
+
+  Usage example:
+
+  ```ts
+  await browser_click({ ref: '@e1', waitUntil: 'domcontentloaded', timeout: 5000 });
+  ```
+
+  Fixes #17397.
+
+### Patch Changes
+
+- Updated dependencies [[`19a8658`](https://github.com/mastra-ai/mastra/commit/19a86589c788ef48bb6c1b0612cc82a201857379), [`a659a77`](https://github.com/mastra-ai/mastra/commit/a659a779bdebe3a52a518c56d2260592d0240fe0), [`3332be9`](https://github.com/mastra-ai/mastra/commit/3332be9701ecd77aba840959d9a1d1ce7aef02d3)]:
+  - @mastra/core@1.38.0-alpha.6
+
 ## 0.3.0-alpha.0
 
 ### Minor Changes

package/dist/index.cjs

@@ -141,7 +141,9 @@ var clickInputSchema = zod.z.object({
   ref: zod.z.string().describe("Element ref from snapshot (e.g., @e5)"),
   button: zod.z.enum(["left", "right", "middle"]).optional().describe("Mouse button (default: left)"),
   clickCount: zod.z.number().optional().describe("Number of clicks (default: 1, use 2 for double-click)"),
-  modifiers: zod.z.array(zod.z.enum(["Alt", "Control", "Meta", "Shift"])).optional().describe("Modifier keys to hold")
+  modifiers: zod.z.array(zod.z.enum(["Alt", "Control", "Meta", "Shift"])).optional().describe("Modifier keys to hold"),
+  waitUntil: zod.z.enum(["load", "domcontentloaded", "networkidle"]).optional().describe("If the click triggers a navigation, wait for this page load state before returning"),
+  timeout: zod.z.number().nonnegative().optional().describe("Timeout in milliseconds for the click and optional waitUntil")
 });
 var typeInputSchema = zod.z.object({
   ref: zod.z.string().describe("Element ref from snapshot"),
@@ -151,13 +153,17 @@ var typeInputSchema = zod.z.object({
 });
 var pressInputSchema = zod.z.object({
   key: zod.z.string().describe("Key to press (e.g., Enter, Tab, Escape, Control+a)"),
-  modifiers: zod.z.array(zod.z.enum(["Alt", "Control", "Meta", "Shift"])).optional().describe("Modifier keys to hold")
+  modifiers: zod.z.array(zod.z.enum(["Alt", "Control", "Meta", "Shift"])).optional().describe("Modifier keys to hold"),
+  waitUntil: zod.z.enum(["load", "domcontentloaded", "networkidle"]).optional().describe("If the key press triggers a navigation, wait for this page load state before returning"),
+  timeout: zod.z.number().nonnegative().optional().describe("Timeout in milliseconds for the optional waitUntil")
 });
 var selectInputSchema = zod.z.object({
   ref: zod.z.string().describe("Select element ref from snapshot"),
   value: zod.z.string().optional().describe("Option value to select"),
   label: zod.z.string().optional().describe("Option label to select"),
-  index: zod.z.number().int().min(0).optional().describe("Option index to select (0-based)")
+  index: zod.z.number().int().min(0).optional().describe("Option index to select (0-based)"),
+  waitUntil: zod.z.enum(["load", "domcontentloaded", "networkidle"]).optional().describe("If the selection triggers a navigation, wait for this page load state before returning"),
+  timeout: zod.z.number().nonnegative().optional().describe("Timeout in milliseconds for the selection and optional waitUntil")
 }).superRefine((data, ctx) => {
   if (data.value === void 0 && data.label === void 0 && data.index === void 0) {
     ctx.addIssue({
@@ -293,7 +299,7 @@ function createBackTool(browser) {
 function createClickTool(browser) {
   return tools.createTool({
     id: BROWSER_TOOLS.CLICK,
-    description: "Click an element using its ref from a snapshot. Use clickCount: 2 for double-click.",
+    description: "Click an element using its ref from a snapshot. Use clickCount: 2 for double-click. Pass waitUntil when the click triggers navigation so the page settles before the next snapshot.",
     inputSchema: clickInputSchema,
     execute: async (input, { agent }) => {
       const threadId = agent?.threadId;
@@ -390,7 +396,7 @@ function createHoverTool(browser) {
 function createPressTool(browser) {
   return tools.createTool({
     id: BROWSER_TOOLS.PRESS,
-    description: "Press a keyboard key (e.g., Enter, Tab, Escape, Control+a).",
+    description: "Press a keyboard key (e.g., Enter, Tab, Escape, Control+a). Pass waitUntil when the keypress triggers navigation (e.g., Enter to submit a form) so the page settles before the next snapshot.",
     inputSchema: pressInputSchema,
     execute: async (input, { agent }) => {
       const threadId = agent?.threadId;
@@ -448,7 +454,7 @@ function createScrollTool(browser) {
 function createSelectTool(browser) {
   return tools.createTool({
     id: BROWSER_TOOLS.SELECT,
-    description: "Select an option from a dropdown by value, label, or index.",
+    description: "Select an option from a dropdown by value, label, or index. Pass waitUntil when the selection triggers navigation so the page settles before the next snapshot.",
     inputSchema: selectInputSchema,
     execute: async (input, { agent }) => {
       const threadId = agent?.threadId;
@@ -1091,12 +1097,15 @@ var AgentBrowser = class extends browser.MastraBrowser {
           "Take a new snapshot to see the current page state and get fresh refs."
         );
       }
+      const timeout = input.timeout ?? this.defaultTimeout;
+      const navigation = input.waitUntil ? page.waitForNavigation({ waitUntil: input.waitUntil, timeout }) : void 0;
       await locator.click({
         button: input.button ?? "left",
         clickCount: input.clickCount ?? 1,
         modifiers: input.modifiers,
-        timeout: this.defaultTimeout
+        timeout
       });
+      await navigation;
       return {
         success: true,
         url: page.url(),
@@ -1165,7 +1174,10 @@ var AgentBrowser = class extends browser.MastraBrowser {
   async press(input, threadId) {
     try {
       const page = await this.getPage(threadId);
+      const timeout = input.timeout ?? this.defaultTimeout;
+      const navigation = input.waitUntil ? page.waitForNavigation({ waitUntil: input.waitUntil, timeout }) : void 0;
       await page.keyboard.press(input.key);
+      await navigation;
       return {
         success: true,
         url: page.url(),
@@ -1193,9 +1205,10 @@ var AgentBrowser = class extends browser.MastraBrowser {
       if (input.value) selectValue.value = input.value;
       if (input.label) selectValue.label = input.label;
       if (input.index !== void 0) selectValue.index = input.index;
-      const selected = await locator.selectOption(selectValue, {
-        timeout: this.defaultTimeout
-      });
+      const timeout = input.timeout ?? this.defaultTimeout;
+      const navigation = input.waitUntil ? page.waitForNavigation({ waitUntil: input.waitUntil, timeout }) : void 0;
+      const selected = await locator.selectOption(selectValue, { timeout });
+      await navigation;
       return {
         success: true,
         selected,