fix: solve merege conflict

Author: bittoby

.gitignore

@@ -63,3 +63,4 @@ resources/prebuilt/cache/
 
 *storybook.log
 storybook-static
+nul

backend/app/agent/prompt.py

@@ -30,12 +30,14 @@
 Your integrated toolkits enable you to:
 
 1. Skills System: You have access to a library of specialized skills that
-   provide expert guidance for specific tasks. When a skill is referenced with
-   double curly braces (e.g., {{pdf}} or {{data-analyzer}}), you should use
-   `list_skills` to discover available skills and `load_skill` to retrieve the
-   skill's full content. Skills contain tested code examples, best practices,
-   and detailed instructions that you MUST follow as your PRIMARY reference when
-   working on tasks that mention them.
+   provide expert guidance for specific tasks. When a task references a skill
+   with double curly braces (e.g., {{pdf}} or {{data-analyzer}}), or when you
+   determine that an available skill is relevant to the task, you MUST use
+   `load_skill` to retrieve the skill's full content BEFORE taking any other
+   action on the task. Skills contain requirements, tested code examples, and
+   best practices that you MUST follow as your PRIMARY reference. Do not begin
+   executing the task until you have loaded and understood the relevant skill
+   instructions.
 
 2. WhatsApp Business Management (WhatsAppToolkit):
    - Send text and template messages to customers via the WhatsApp Business
@@ -151,12 +153,14 @@
 <capabilities>
 Your capabilities include:
 - **Skills System**: You have access to a library of specialized skills that
-  provide expert guidance for specific tasks. When a skill is referenced with
-  double curly braces (e.g., {{pdf}} or {{data-analyzer}}), you should use
-  `list_skills` to discover available skills and `load_skill` to retrieve the
-  skill's full content. Skills contain tested code examples, best practices,
-  and detailed instructions that you MUST follow as your PRIMARY reference when
-  working on tasks that mention them.
+  provide expert guidance for specific tasks. When a task references a skill
+  with double curly braces (e.g., {{pdf}} or {{data-analyzer}}), or when you
+  determine that an available skill is relevant to the task, you MUST use
+  `load_skill` to retrieve the skill's full content BEFORE taking any other
+  action on the task. Skills contain requirements, tested code examples, and
+  best practices that you MUST follow as your PRIMARY reference. Do not begin
+  executing the task until you have loaded and understood the relevant skill
+  instructions.
 - Video & Audio Analysis:
     - Download videos from URLs for analysis.
     - Transcribe speech from audio files to text with high accuracy
@@ -280,12 +284,14 @@
 Your capabilities include:
 - You can use ScreenshotToolkit to read image with given path.
 - **Skills System**: You have access to a library of specialized skills that
-  provide expert guidance for specific tasks. When a skill is referenced with
-  double curly braces (e.g., {{pdf}} or {{data-analyzer}}), you should use
-  `list_skills` to discover available skills and `load_skill` to retrieve the
-  skill's full content. Skills contain tested code examples, best practices,
-  and detailed instructions that you MUST follow as your PRIMARY reference when
-  working on tasks that mention them.
+  provide expert guidance for specific tasks. When a task references a skill
+  with double curly braces (e.g., {{pdf}} or {{data-analyzer}}), or when you
+  determine that an available skill is relevant to the task, you MUST use
+  `load_skill` to retrieve the skill's full content BEFORE taking any other
+  action on the task. Skills contain requirements, tested code examples, and
+  best practices that you MUST follow as your PRIMARY reference. Do not begin
+  executing the task until you have loaded and understood the relevant skill
+  instructions.
 - Document Reading:
     - Read and understand the content of various file formats including
         - PDF (.pdf)
@@ -438,12 +444,14 @@
 Your capabilities are extensive and powerful:
 - You can use ScreenshotToolkit to read image with given path.
 - **Skills System**: You have access to a library of specialized skills that
-  provide expert guidance for specific tasks. When a skill is referenced with
-  double curly braces (e.g., {{pdf}} or {{data-analyzer}}), you should use
-  `list_skills` to discover available skills and `load_skill` to retrieve the
-  skill's full content. Skills contain tested code examples, best practices,
-  and detailed instructions that you MUST follow as your PRIMARY reference when
-  working on tasks that mention them.
+  provide expert guidance for specific tasks. When a task references a skill
+  with double curly braces (e.g., {{pdf}} or {{data-analyzer}}), or when you
+  determine that an available skill is relevant to the task, you MUST use
+  `load_skill` to retrieve the skill's full content BEFORE taking any other
+  action on the task. Skills contain requirements, tested code examples, and
+  best practices that you MUST follow as your PRIMARY reference. Do not begin
+  executing the task until you have loaded and understood the relevant skill
+  instructions.
 - **Unrestricted Code Execution**: You can write and execute code in any
   language to solve a task. You MUST first save your code to a file (e.g.,
   `script.py`) and then run it from the terminal (e.g.,
@@ -614,12 +622,14 @@
 Your capabilities include:
 - You can use ScreenshotToolkit to read image with given path.
 - **Skills System**: You have access to a library of specialized skills that
-  provide expert guidance for specific tasks. When a skill is referenced with
-  double curly braces (e.g., {{pdf}} or {{data-analyzer}}), you should use
-  `list_skills` to discover available skills and `load_skill` to retrieve the
-  skill's full content. Skills contain tested code examples, best practices,
-  and detailed instructions that you MUST follow as your PRIMARY reference when
-  working on tasks that mention them.
+  provide expert guidance for specific tasks. When a task references a skill
+  with double curly braces (e.g., {{pdf}} or {{data-analyzer}}), or when you
+  determine that an available skill is relevant to the task, you MUST use
+  `load_skill` to retrieve the skill's full content BEFORE taking any other
+  action on the task. Skills contain requirements, tested code examples, and
+  best practices that you MUST follow as your PRIMARY reference. Do not begin
+  executing the task until you have loaded and understood the relevant skill
+  instructions.
 - Search and get information from the web using the search tools.
 - Use the rich browser related toolset to investigate websites.
 - Use the terminal tools to perform local operations. **IMPORTANT:** Before the

backend/app/controller/tool_controller.py

@@ -14,6 +14,7 @@
 
 import logging
 import os
+import shutil
 import time
 
 from fastapi import APIRouter, HTTPException
@@ -712,16 +713,39 @@ def is_port_in_use(port):
                 f"Electron browser script not found: {electron_script_path}"
             )
 
-        electron_cmd = "npx"
-        electron_args = [
-            electron_cmd,
+        # Resolve npx path for Windows compatibility.
+        # On Windows, subprocess.Popen uses CreateProcess which cannot
+        # execute .cmd files directly. We resolve the full path and
+        # invoke via cmd.exe.
+        npx_cmd = None
+        if os.name == "nt":
+            eigent_npx = os.path.expanduser("~/.eigent/bin/npx.cmd")
+            if os.path.exists(eigent_npx):
+                npx_cmd = eigent_npx
+        if not npx_cmd:
+            npx_cmd = shutil.which("npx") or shutil.which("npx.cmd")
+        if not npx_cmd:
+            if os.name == "nt":
+                raise FileNotFoundError(
+                    "npx not found. Please ensure Node.js is installed and npx is on your PATH."
+                )
+            npx_cmd = "npx"
+
+        base_args = [
+            npx_cmd,
             "electron",
             electron_script_path,
             user_data_dir,
             str(cdp_port),
             "https://www.google.com",
         ]
 
+        # On Windows, wrap with cmd.exe so .cmd execution is reliable
+        if os.name == "nt":
+            electron_args = ["cmd.exe", "/d", "/s", "/c"] + base_args
+        else:
+            electron_args = base_args
+
         # Get the app's directory to run npx in the right context
         app_dir = os.path.dirname(
             os.path.dirname(os.path.dirname(os.path.dirname(__file__)))
@@ -736,10 +760,17 @@ def is_port_in_use(port):
         logger.info(f"[PROFILE USER LOGIN] userData path: {user_data_dir}")
         logger.info(f"[PROFILE USER LOGIN] Electron args: {electron_args}")
 
+        # Ensure ~/.eigent/bin is on PATH for the spawned process
+        env = os.environ.copy()
+        eigent_bin = os.path.expanduser("~/.eigent/bin")
+        if os.path.isdir(eigent_bin):
+            env["PATH"] = eigent_bin + os.pathsep + env.get("PATH", "")
+
         # Start process and capture output in real-time
         process = subprocess.Popen(
             electron_args,
             cwd=app_dir,
+            env=env,
             stdout=subprocess.PIPE,
             stderr=subprocess.STDOUT,  # Redirect stderr to stdout
             text=True,
@@ -793,6 +824,20 @@ async def log_electron_output():
         )
 
 
+@router.get("/browser/status", name="browser status")
+async def browser_status():
+    """Check if the login browser is currently open."""
+    import socket
+
+    cdp_port = 9223
+
+    def is_port_in_use(port):
+        with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
+            return s.connect_ex(("localhost", port)) == 0
+
+    return {"is_open": is_port_in_use(cdp_port)}
+
+
 @router.get("/browser/cookies", name="list cookie domains")
 async def list_cookie_domains(search: str = None):
     """

backend/app/model/model_platform.py

@@ -20,6 +20,7 @@
     "z.ai": "openai-compatible-model",
     "ModelArk": "openai-compatible-model",
     "grok": "openai-compatible-model",
+    "llama.cpp": "openai-compatible-model",
 }
 
 

backend/tests/app/model/test_model_platform.py

@@ -26,6 +26,7 @@ def test_normalize_model_platform_maps_known_aliases():
     assert normalize_model_platform("grok") == "openai-compatible-model"
     assert normalize_model_platform("z.ai") == "openai-compatible-model"
     assert normalize_model_platform("ModelArk") == "openai-compatible-model"
+    assert normalize_model_platform("llama.cpp") == "openai-compatible-model"
 
 
 def test_normalize_model_platform_keeps_non_alias_unchanged():
@@ -43,7 +44,7 @@ class _Model(BaseModel):
         optional_model_platform: NormalizedOptionalModelPlatform = None
 
     item = _Model(
-        model_platform="grok",
+        model_platform="llama.cpp",
         optional_model_platform="ModelArk",
     )
 

docs/core/models/local-model.md

@@ -38,6 +38,11 @@ print(f"Server started on http://localhost:{port}")
 ollama pull qwen2.5:7b
 ```
 
+```bash
+# LLaMA.cpp server https://github.com/ggml-org/llama.cpp/tree/master/tools/server
+./llama-server -m /path/to/model.gguf --host 0.0.0.0 --port 8080
+```
+
 2. Setting your model
 
 ![set_local_model](/docs/images/models_local_model.png)

docs/get_started/quick_start.md

@@ -173,7 +173,7 @@ Eigent can run in two modes. Your choice here affects how you are billed and wha
 - **Cloud Version:** We provide pre-configured, state-of-the-art models, including GPT-4.1, GPT-4.1 mini and Gemini 2.5 Pro. Using these models is the easiest way to get started and will be billed to your account based on usage (credits).
 - **Self-hosted Version:** You can connect your own models.
   - **Cloud Models:** Connect your personal accounts from providers like OpenAI, Anthropic, Qwen, Deepseek and Azure by entering your own API key.
-  - **Local Models:** For advanced users, you can run models locally using Ollama, vLLM, or SGLang.
+  - **Local Models:** For advanced users, you can run models locally using Ollama, vLLM, SGLang, LM Studio, or LLaMA.cpp server.
 
 ### **MCP Servers**
 

electron/main/index.ts

@@ -693,9 +693,10 @@ function registerIpcHandlers() {
   // Launch CDP browser with automatic port assignment
   ipcMain.handle('launch-cdp-browser', async () => {
     try {
-      // 1. Find available port (9223–9300) by checking no CDP browser is listening
+      // 1. Find available port (9224–9300) by checking no CDP browser is listening
+      // Port 9223 is reserved for the login browser
       let port: number | null = null;
-      for (let p = 9223; p < 9300; p++) {
+      for (let p = 9224; p < 9300; p++) {
         if (
           !cdp_browser_pool.some((b) => b.port === p) &&
           !(await isCdpPortAlive(p))
@@ -705,7 +706,7 @@ function registerIpcHandlers() {
         }
       }
       if (port === null) {
-        return { success: false, error: 'No available port in 9223-9299' };
+        return { success: false, error: 'No available port in 9224-9299' };
       }
 
       // 2. Find Playwright Chromium executable

package.json

@@ -70,6 +70,7 @@
     "@radix-ui/react-toggle-group": "^1.1.10",
     "@radix-ui/react-tooltip": "^1.2.7",
     "@stackframe/react": "file:package/@stackframe/react",
+    "@tanstack/react-query": "^5.90.19",
     "@xterm/addon-fit": "^0.10.0",
     "@xterm/addon-web-links": "^0.11.0",
     "@xterm/xterm": "^5.5.0",
@@ -81,6 +82,7 @@
     "clsx": "^2.1.1",
     "cmdk": "^1.1.1",
     "csv-parser": "^3.2.0",
+    "date-fns": "^3.6.0",
     "dompurify": "^3.2.7",
     "electron-log": "^5.4.0",
     "electron-updater": "^6.3.9",
@@ -101,6 +103,7 @@
     "next-themes": "^0.4.6",
     "papaparse": "^5.5.3",
     "postprocessing": "^6.37.8",
+    "react-day-picker": "^9.13.0",
     "react-markdown": "^10.1.0",
     "react-resizable-panels": "^3.0.4",
     "react-router-dom": "^7.6.0",

server/.env.example

@@ -1,7 +1,31 @@
 debug=false
 url_prefix=/api
 secret_key=postgres
-database_url=postgresql://postgres:123456@localhost:5432/postgres
 # Chat Share Secret Key
 CHAT_SHARE_SECRET_KEY=put-your-secret-key-here
 CHAT_SHARE_SALT=put-your-encode-salt-here
+
+# Configuration if not running in docker
+# database_url=postgresql://postgres:123456@localhost:5432/eigent
+# redis_url=redis://localhost:6379/0
+# celery_broker_url=redis://localhost:6379/0
+# celery_result_url=redis://localhost:6379/0
+# SESSION_REDIS_URL=redis://localhost:6379/1
+
+# Trigger Schedule Poller Configuration
+# ENABLE_TRIGGER_SCHEDULE_POLLER_TASK: Enable/disable scheduled trigger polling
+ENABLE_TRIGGER_SCHEDULE_POLLER_TASK=true
+# TRIGGER_SCHEDULE_POLLER_INTERVAL: Polling interval in minutes
+TRIGGER_SCHEDULE_POLLER_INTERVAL=1
+# TRIGGER_SCHEDULE_POLLER_BATCH_SIZE: Number of triggers to fetch per poll
+TRIGGER_SCHEDULE_POLLER_BATCH_SIZE=100
+# TRIGGER_SCHEDULE_MAX_DISPATCH_PER_TICK: Max triggers to dispatch per tick (0 = unlimited)
+TRIGGER_SCHEDULE_MAX_DISPATCH_PER_TICK=0
+# ENABLE_EXECUTION_TIMEOUT_CHECKER: Enable/disable execution timeout checking
+ENABLE_EXECUTION_TIMEOUT_CHECKER=true
+# EXECUTION_TIMEOUT_CHECKER_INTERVAL: check_execution_timeouts interval in minutes
+EXECUTION_TIMEOUT_CHECKER_INTERVAL=1
+# EXECUTION_PENDING_TIMEOUT_SECONDS: Timeout for pending executions (default 60 seconds)
+EXECUTION_PENDING_TIMEOUT_SECONDS=60
+# EXECUTION_RUNNING_TIMEOUT_SECONDS: Timeout for running executions (default 600 seconds / 10 minutes)
+EXECUTION_RUNNING_TIMEOUT_SECONDS=600