@hypothesi/tauri-mcp-server 0.1.2 → 0.2.0

package/dist/driver/webview-interactions.js

@@ -1,10 +1,20 @@
 import { z } from 'zod';
-import { executeInWebview, captureScreenshot, getConsoleLogs as getConsoleLogsFromCapture } from './webview-executor.js';
+import { executeInWebview, executeInWebviewWithContext, captureScreenshot, getConsoleLogs as getConsoleLogsFromCapture, } from './webview-executor.js';
 import { SCRIPTS, buildScript, buildTypeScript, buildKeyEventScript } from './scripts/index.js';
 // ============================================================================
+// Base Schema for Window Targeting
+// ============================================================================
+/**
+ * Base schema mixin for tools that can target a specific window.
+ * All webview tools extend this to support multi-window applications.
+ */
+export const WindowTargetSchema = z.object({
+    windowId: z.string().optional().describe('Window label to target (defaults to "main")'),
+});
+// ============================================================================
 // Schemas
 // ============================================================================
-export const InteractSchema = z.object({
+export const InteractSchema = WindowTargetSchema.extend({
     action: z.enum(['click', 'double-click', 'long-press', 'scroll', 'swipe'])
         .describe('Type of interaction to perform'),
     selector: z.string().optional().describe('CSS selector for the element to interact with'),
@@ -14,17 +24,16 @@ export const InteractSchema = z.object({
         .describe('Duration in ms for long-press or swipe (default: 500ms for long-press, 300ms for swipe)'),
     scrollX: z.number().optional().describe('Horizontal scroll amount in pixels (positive = right)'),
     scrollY: z.number().optional().describe('Vertical scroll amount in pixels (positive = down)'),
-    // Swipe-specific parameters
     fromX: z.number().optional().describe('Starting X coordinate for swipe'),
     fromY: z.number().optional().describe('Starting Y coordinate for swipe'),
     toX: z.number().optional().describe('Ending X coordinate for swipe'),
     toY: z.number().optional().describe('Ending Y coordinate for swipe'),
 });
-export const ScreenshotSchema = z.object({
+export const ScreenshotSchema = WindowTargetSchema.extend({
     format: z.enum(['png', 'jpeg']).optional().default('png').describe('Image format'),
     quality: z.number().min(0).max(100).optional().describe('JPEG quality (0-100, only for jpeg format)'),
 });
-export const KeyboardSchema = z.object({
+export const KeyboardSchema = WindowTargetSchema.extend({
     action: z.enum(['type', 'press', 'down', 'up'])
         .describe('Keyboard action type: "type" for typing text into an element, "press/down/up" for key events'),
     selector: z.string().optional().describe('CSS selector for element to type into (required for "type" action)'),
@@ -32,29 +41,29 @@ export const KeyboardSchema = z.object({
     key: z.string().optional().describe('Key to press (required for "press/down/up" actions, e.g., "Enter", "a", "Escape")'),
     modifiers: z.array(z.enum(['Control', 'Alt', 'Shift', 'Meta'])).optional().describe('Modifier keys to hold'),
 });
-export const WaitForSchema = z.object({
+export const WaitForSchema = WindowTargetSchema.extend({
     type: z.enum(['selector', 'text', 'ipc-event']).describe('What to wait for'),
     value: z.string().describe('Selector, text content, or IPC event name to wait for'),
     timeout: z.number().optional().default(5000).describe('Timeout in milliseconds (default: 5000ms)'),
 });
-export const GetStylesSchema = z.object({
+export const GetStylesSchema = WindowTargetSchema.extend({
     selector: z.string().describe('CSS selector for element(s) to get styles from'),
     properties: z.array(z.string()).optional().describe('Specific CSS properties to retrieve. If omitted, returns all computed styles'),
     multiple: z.boolean().optional().default(false)
         .describe('Whether to get styles for all matching elements (true) or just the first (false)'),
 });
-export const ExecuteJavaScriptSchema = z.object({
+export const ExecuteJavaScriptSchema = WindowTargetSchema.extend({
     script: z.string().describe('JavaScript code to execute in the webview context'),
     args: z.array(z.unknown()).optional().describe('Arguments to pass to the script'),
 });
-export const FocusElementSchema = z.object({
+export const FocusElementSchema = WindowTargetSchema.extend({
     selector: z.string().describe('CSS selector for element to focus'),
 });
-export const FindElementSchema = z.object({
+export const FindElementSchema = WindowTargetSchema.extend({
     selector: z.string(),
     strategy: z.enum(['css', 'xpath', 'text']).default('css'),
 });
-export const GetConsoleLogsSchema = z.object({
+export const GetConsoleLogsSchema = WindowTargetSchema.extend({
     filter: z.string().optional().describe('Regex or keyword to filter logs'),
     since: z.string().optional().describe('ISO timestamp to filter logs since'),
 });
@@ -62,10 +71,10 @@ export const GetConsoleLogsSchema = z.object({
 // Implementation Functions
 // ============================================================================
 export async function interact(options) {
-    const { action, selector, x, y, duration, scrollX, scrollY, fromX, fromY, toX, toY } = options;
+    const { action, selector, x, y, duration, scrollX, scrollY, fromX, fromY, toX, toY, windowId } = options;
     // Handle swipe action separately since it has different logic
     if (action === 'swipe') {
-        return performSwipe(fromX, fromY, toX, toY, duration);
+        return performSwipe({ fromX, fromY, toX, toY, duration, windowId });
     }
     const script = buildScript(SCRIPTS.interact, {
         action,
@@ -77,31 +86,34 @@ export async function interact(options) {
         scrollY: scrollY ?? 0,
     });
     try {
-        return await executeInWebview(script);
+        return await executeInWebview(script, windowId);
     }
     catch (error) {
         const message = error instanceof Error ? error.message : String(error);
         throw new Error(`Interaction failed: ${message}`);
     }
 }
-async function performSwipe(fromX, fromY, toX, toY, duration = 300) {
+async function performSwipe(options) {
+    const { fromX, fromY, toX, toY, duration = 300, windowId } = options;
     if (fromX === undefined || fromY === undefined || toX === undefined || toY === undefined) {
         throw new Error('Swipe action requires fromX, fromY, toX, and toY coordinates');
     }
     const script = buildScript(SCRIPTS.swipe, { fromX, fromY, toX, toY, duration });
     try {
-        return await executeInWebview(script);
+        return await executeInWebview(script, windowId);
     }
     catch (error) {
         const message = error instanceof Error ? error.message : String(error);
         throw new Error(`Swipe failed: ${message}`);
     }
 }
-export async function screenshot(quality, format = 'png') {
+export async function screenshot(options = {}) {
+    const { quality, format = 'png', windowId } = options;
     // Use the native screenshot function from webview-executor
-    return captureScreenshot(format, quality);
+    return captureScreenshot({ format, quality, windowId });
 }
-export async function keyboard(action, selectorOrKey, textOrModifiers, modifiers) {
+export async function keyboard(options) {
+    const { action, selectorOrKey, textOrModifiers, modifiers, windowId } = options;
     // Handle the different parameter combinations based on action
     if (action === 'type') {
         const selector = selectorOrKey;
@@ -111,7 +123,7 @@ export async function keyboard(action, selectorOrKey, textOrModifiers, modifiers
         }
         const script = buildTypeScript(selector, text);
         try {
-            return await executeInWebview(script);
+            return await executeInWebview(script, windowId);
         }
         catch (error) {
             const message = error instanceof Error ? error.message : String(error);
@@ -126,38 +138,41 @@ export async function keyboard(action, selectorOrKey, textOrModifiers, modifiers
     }
     const script = buildKeyEventScript(action, key, mods || []);
     try {
-        return await executeInWebview(script);
+        return await executeInWebview(script, windowId);
     }
     catch (error) {
         const message = error instanceof Error ? error.message : String(error);
         throw new Error(`Keyboard action failed: ${message}`);
     }
 }
-export async function waitFor(type, value, timeout = 5000) {
+export async function waitFor(options) {
+    const { type, value, timeout = 5000, windowId } = options;
     const script = buildScript(SCRIPTS.waitFor, { type, value, timeout });
     try {
-        return await executeInWebview(script);
+        return await executeInWebview(script, windowId);
     }
     catch (error) {
         const message = error instanceof Error ? error.message : String(error);
         throw new Error(`Wait failed: ${message}`);
     }
 }
-export async function getStyles(selector, properties, multiple = false) {
+export async function getStyles(options) {
+    const { selector, properties, multiple = false, windowId } = options;
     const script = buildScript(SCRIPTS.getStyles, {
         selector,
         properties: properties || [],
         multiple,
     });
     try {
-        return await executeInWebview(script);
+        return await executeInWebview(script, windowId);
     }
     catch (error) {
         const message = error instanceof Error ? error.message : String(error);
         throw new Error(`Get styles failed: ${message}`);
     }
 }
-export async function executeJavaScript(script, args) {
+export async function executeJavaScript(options) {
+    const { script, args, windowId } = options;
     // If args are provided, we need to inject them into the script context
     const wrappedScript = args && args.length > 0
         ? `
@@ -168,18 +183,26 @@ export async function executeJavaScript(script, args) {
       `
         : script;
     try {
-        const result = await executeInWebview(wrappedScript);
-        return result;
+        const { result, windowLabel, warning } = await executeInWebviewWithContext(wrappedScript, windowId);
+        // Build response with window context
+        let response = result;
+        if (warning) {
+            response = `⚠️ ${warning}\n\n${response}`;
+        }
+        // Add window info footer for clarity
+        response += `\n\n[Executed in window: ${windowLabel}]`;
+        return response;
     }
     catch (error) {
         const message = error instanceof Error ? error.message : String(error);
         throw new Error(`JavaScript execution failed: ${message}`);
     }
 }
-export async function focusElement(selector) {
+export async function focusElement(options) {
+    const { selector, windowId } = options;
     const script = buildScript(SCRIPTS.focus, { selector });
     try {
-        return await executeInWebview(script);
+        return await executeInWebview(script, windowId);
     }
     catch (error) {
         const message = error instanceof Error ? error.message : String(error);
@@ -189,10 +212,11 @@ export async function focusElement(selector) {
 /**
  * Find an element using various selector strategies.
  */
-export async function findElement(selector, strategy) {
+export async function findElement(options) {
+    const { selector, strategy, windowId } = options;
     const script = buildScript(SCRIPTS.findElement, { selector, strategy });
     try {
-        return await executeInWebview(script);
+        return await executeInWebview(script, windowId);
     }
     catch (error) {
         const message = error instanceof Error ? error.message : String(error);
@@ -202,7 +226,8 @@ export async function findElement(selector, strategy) {
 /**
  * Get console logs from the webview.
  */
-export async function getConsoleLogs(filter, since) {
+export async function getConsoleLogs(options = {}) {
+    const { filter, since } = options;
     try {
         return await getConsoleLogsFromCapture(filter, since);
     }

package/dist/index.js

@@ -1,13 +1,14 @@
 #!/usr/bin/env node
 import { Server } from '@modelcontextprotocol/sdk/server/index.js';
 import { StdioServerTransport } from '@modelcontextprotocol/sdk/server/stdio.js';
-import { CallToolRequestSchema, ListToolsRequestSchema, } from '@modelcontextprotocol/sdk/types.js';
+import { CallToolRequestSchema, ListToolsRequestSchema, ListPromptsRequestSchema, GetPromptRequestSchema, } from '@modelcontextprotocol/sdk/types.js';
 import { zodToJsonSchema } from 'zod-to-json-schema';
 import { readFileSync } from 'fs';
 import { fileURLToPath } from 'url';
 import { dirname, join } from 'path';
-// Import the single source of truth for all tools
+// Import the single source of truth for all tools and prompts
 import { TOOLS, TOOL_MAP } from './tools-registry.js';
+import { PROMPTS, PROMPT_MAP } from './prompts-registry.js';
 /* eslint-disable no-process-exit */
 // Read version from package.json
 const currentDir = dirname(fileURLToPath(import.meta.url));
@@ -20,6 +21,7 @@ const server = new Server({
 }, {
     capabilities: {
         tools: {},
+        prompts: {},
     },
 });
 // Handle connection errors gracefully - don't crash on broken pipe
@@ -45,6 +47,7 @@ server.setRequestHandler(ListToolsRequestSchema, async () => {
                 name: tool.name,
                 description: tool.description,
                 inputSchema: zodToJsonSchema(tool.schema),
+                annotations: tool.annotations,
             };
         }),
     };
@@ -67,6 +70,30 @@ server.setRequestHandler(CallToolRequestSchema, async (request) => {
         };
     }
 });
+// Prompt list handler - generated from registry
+server.setRequestHandler(ListPromptsRequestSchema, async () => {
+    return {
+        prompts: PROMPTS.map((prompt) => {
+            return {
+                name: prompt.name,
+                description: prompt.description,
+                arguments: prompt.arguments,
+            };
+        }),
+    };
+});
+// Get prompt handler - returns prompt messages for a specific prompt
+server.setRequestHandler(GetPromptRequestSchema, async (request) => {
+    const prompt = PROMPT_MAP.get(request.params.name);
+    if (!prompt) {
+        throw new Error(`Unknown prompt: ${request.params.name}`);
+    }
+    const args = (request.params.arguments || {});
+    return {
+        description: prompt.description,
+        messages: prompt.handler(args),
+    };
+});
 // Start server
 async function main() {
     const transport = new StdioServerTransport();

package/dist/prompts-registry.js

@@ -0,0 +1,54 @@
+/**
+ * Single source of truth for all MCP prompt definitions
+ * Prompts are user-controlled templates that appear as slash commands in MCP clients
+ */
+const FIX_WEBVIEW_ERRORS_PROMPT = `I need help finding and fixing JavaScript errors in my Tauri app's webview.
+
+Please follow these steps:
+
+1. **Start a session** - Use \`tauri_driver_session\` with action "start" to connect to the running Tauri app
+
+2. **Get console logs** - Use \`tauri_driver_get_console_logs\` to retrieve any JavaScript errors or warnings from the webview console
+
+3. **Analyze the errors** - Look at the error messages, stack traces, and identify:
+   - What type of error it is (TypeError, ReferenceError, SyntaxError, etc.)
+   - Which file and line number the error originates from
+   - What the root cause might be
+
+4. **Find the source code** - Use code search or file reading tools to locate the problematic code in my project
+
+5. **Propose a fix** - Explain what's wrong and suggest a concrete fix for each error found
+
+6. **Stop the session** - Use \`tauri_driver_session\` with action "stop" to clean up
+
+If no errors are found, let me know the app is running cleanly.
+
+If the session fails to start, help me troubleshoot the connection (is the app running? is the MCP bridge plugin installed?).`;
+/**
+ * Complete registry of all available prompts
+ */
+export const PROMPTS = [
+    {
+        name: 'fix-webview-errors',
+        description: '[Tauri Apps Only] Find and fix JavaScript errors in a running Tauri app. ' +
+            'Use ONLY for Tauri projects (with src-tauri/ and tauri.conf.json). ' +
+            'For browser debugging, use Chrome DevTools MCP instead. ' +
+            'For Electron apps, this prompt will NOT work.',
+        arguments: [],
+        handler: () => {
+            return [
+                {
+                    role: 'user',
+                    content: {
+                        type: 'text',
+                        text: FIX_WEBVIEW_ERRORS_PROMPT,
+                    },
+                },
+            ];
+        },
+    },
+];
+/**
+ * Create a Map for fast prompt lookup by name
+ */
+export const PROMPT_MAP = new Map(PROMPTS.map((prompt) => { return [prompt.name, prompt]; }));