[Bug Report]: The task stopped before it was completed.

[iamtornado]

Version

0.3.0

Issue Type

• Select a issue type 👇
• Agent TARS Web UI (@agent-tars/web-ui)
• Agent TARS CLI (@agent-tars/server)
• Agent TARS Server (@agent-tars/server)
• Agent TARS (@agent-tars/core)
• MCP Agent (@tarko/mcp-agent)
• Agent Kernel (@tarko/agent)
• Other (please specify in description)

Model Provider

• Select a model provider 👇
• Volcengine
• Anthropic
• OpenAI
• Azure OpenAI
• Other (please specify in description)

Problem Description

The task stopped before it was completed.
The same issue occurs with the web UI.
OS is windows11 24H2.

commad is:

agent-tars run --input "Tell me the top 5 most popular projects on ProductHunt today" `
--provider volcengine `
--model doubao-1-5-thinking-vision-pro-250428 `
--apiKey {my-api-key} `
--logLevel debug `
--search.provider bing_search `
--planner.enable `
--debug

Error Logs

+------------------------------------------------+
| |
| █████ ██████ ███████ ███ ██ ████████ |
| ██ ██ ██ ██ ████ ██ ██ |
| ███████ ██ ██ █████ ██ ██ ██ ██ |
| ██ ██ ██ ██ ██ ██ ██ ██ ██ |
| ██ ██ ███████ ███████ ██ ████ ██ |
| |
| ████████ █████ ██████ ███████ |
| ██ ██ ██ ██ ██ ██ |
| ██ ███████ ██████ ███████ |
| ██ ██ ██ ██ ██ ██ |
| ██ ██ ██ ██ ██ ███████ |
| |
| Agent TARS v0.3.0 |
| |
| An open-source Multimodal AI Agent |
| |
| https://agent-tars.com |
| |
+------------------------------------------------+

AgentCLI Adding CLI config paths: [undefined]
AgentCLI Config search paths: [undefined]
AgentCLI Loading configuration...
AgentCLI Final merged config keys: []
AgentCLI Deprecated CLI options detected: provider, apiKey. Consider using config file format.
AgentCLI Configuration loaded successfully
AgentCLI Model: volcengine (doubao-1-5-thinking-vision-pro-250428)
AgentCLI Server: port 8888, storage sqlite
AgentCLI Using agent: undefined
Setting up workspace static server at: C:\Users\116823
AgentCLI Using static files from: C:\Users\116823\AppData\Roaming\npm\node_modules@agent-tars\cli\node_modules@tarko\agent-ui-builder\static

+------------------------------------------------------------------+
| |
| 🎉 @agent-tars/core is available at: http://localhost:8888 |
| |
| 📁 Workspace: ~ |
| |
| 🤖 Model: volcengine | doubao-1-5-thinking-vision-pro-250428 |
| |
+------------------------------------------------------------------+

AgentTARS:LocalEnvironment:BrowserManager Browser manager initialized (browser not launched yet)
EventStream EventStream initialized with 7 initial events
EventStream EventStream initialized with options: {
maxEvents: 1000,
autoTrim: true,
initialEvents: [
{
id: '99cd6515-0621-4d00-9b1e-a144297d6256',
type: 'user_message',
timestamp: 1762416451020,
content: 'Tell me the top 5 most popular projects on ProductHunt today'
},
{
id: '33648217-87ce-44c6-bb37-a8d0536ee789',
type: 'agent_run_start',
timestamp: 1762416451019,
sessionId: 'G4cwejqq64pORrSzHftlC',
runOptions: [Object],
provider: 'volcengine',
model: 'doubao-1-5-thinking-vision-pro-250428',
agentName: '@agent-tars/core'
},
{
id: '61adfc6a-10d9-4ebb-840a-58ce09fd9815',
type: 'assistant_message',
timestamp: 1762416454291,
content: '用户现在需要获取ProductHunt今日最受欢迎的前5个项目，首先用web_search工具，查询要简洁，所以构造query为“ProductHunt today top projects”，count设为5。',
rawContent: '',
toolCalls: [Array],
finishReason: 'tool_calls',
messageId: 'msg_1762416452578_f61j1umq'
},
{
id: '1cd5664b-e20c-47da-8dc8-bffe66e86c0b',
type: 'tool_call',
timestamp: 1762416454297,
toolCallId: 'call_voxriufmyeheob6uw8lkfzle',
name: 'web_search',
arguments: [Object],
startTime: 1762416454296,
tool: [Object]
},
{
id: 'd0b71ec2-fd8e-4aad-b85e-2b615e7391af',
type: 'tool_result',
timestamp: 1762416457214,
toolCallId: 'call_voxriufmyeheob6uw8lkfzle',
name: 'web_search',
content: [Array],
elapsedMs: 2916
},
{
id: '353523d8-3380-497e-818c-4c3e68516590',
type: 'assistant_message',
timestamp: 1762416459874,
content: '现在需要访问ProductHunt的主页来获取今日热门项目信息，所以选择browser_navigate工具，导航到ProductHunt的官网https://www.producthunt.com/ 。这样可以进入页面后进一步提取内容。[{"name":"browser_navigate","parameters":{"url":"https://www.producthunt.com/"}}]',
rawContent: '',
finishReason: 'stop',
messageId: 'msg_1762416458228_4dwod3tn'
},
{
id: '8bbeaf1e-1f61-4a48-9414-7e75ea421f70',
type: 'agent_run_end',
timestamp: 1762416459881,
sessionId: 'G4cwejqq64pORrSzHftlC',
iterations: 2,
elapsedMs: 8863,
status: 'idle'
}
]
}
EventStream Subscribed to events (total subscribers: 1)
EventStream Subscribed to events (total subscribers: 2)
EventStream Subscribed to events (total subscribers: 3)
AgentTARS:LocalEnvironment:BrowserManager Creating browser instance (not launched yet)
AgentTARS:LocalEnvironment:BrowserManager:LocalBrowser Browser Options: {
logger: ConsoleLogger {
prefix: 'AgentTARS:LocalEnvironment:BrowserManager:LocalBrowser',
lastPrefixColor: null,
level: 1
}
}
ToolManager [Tool] Registered: web_search | Description: "⚠️ SEARCH QUERY LENGTH LIMIT: 7 WORDS MAXIMUM ⚠️

Search the web for information. For best results:

1. Use CONCISE queries (3-5 words ideal)
2. Include only ESSENTIAL keywords, not full questions
3. For complex topics, use multiple simple searches instead of one long query
4. Focus on specific terms that will appear on relevant pages"
   ToolManager [Tool] Registered: browser_vision_control | Description: "A browser operation tool based on visual understanding, perform the next action to complete the task.

Action Space

click(point='x1 y1') - Click at the specified coordinates
left_double(point='x1 y1') - Double-click at the specified coordinates
right_single(point='x1 y1') - Right-click at the specified coordinates
drag(start_point='x1 y1', end_point='x2 y2') - Drag from start to end point
hotkey(key='ctrl c') - Press keyboard shortcut (use space to separate keys, lowercase)
type(content='xxx') - Type text content (use ', ", and \n for special characters)
scroll(point='x1 y1', direction='down or up or right or left') - Scroll in specified direction
wait() - Wait 5 seconds and take a screenshot to check for changes

Note

• Follow user language in in thought part.
• Describe your thought in step part.
• Describe your action in Step part.
• Extract the data your see in pageData part.
• This tool is for operational tasks, not for collect information.
  "
  ToolManager [Tool] Registered: browser_navigate | Description: "[browser] Navigate to a URL"
  ToolManager [Tool] Registered: browser_go_back | Description: "[browser] Go back to the previous page, or close tab if no history exists"
  ToolManager [Tool] Registered: browser_go_forward | Description: "[browser] Go forward to the next page"
  ToolManager [Tool] Registered: browser_refresh | Description: "[browser] Refresh the current page"
  ToolManager [Tool] Registered: browser_get_markdown | Description: "[browser] Get the content of the current page as markdown with pagination support"
  ToolManager [Tool] Registered: browser_screenshot | Description: "[browser] Take a screenshot of the current page or a specific area. this tool SHOULD NOT be called unless the user requests an explicit call."
  ToolManager [Tool] Registered: browser_click | Description: "[browser] Click an element on the page, before using the tool, use browser_get_clickable_elements to get the index of the element, but not call browser_get_clickable_elements multiple times"
  ToolManager [Tool] Registered: browser_form_input_fill | Description: "[browser] Fill out an input field, before using the tool, Either 'index' or 'selector' must be provided"
  ToolManager [Tool] Registered: browser_select | Description: "[browser] Select an element on the page with index, Either 'index' or 'selector' must be provided"
  ToolManager [Tool] Registered: browser_hover | Description: "[browser] Hover an element on the page, Either 'index' or 'selector' must be provided"
  ToolManager [Tool] Registered: browser_evaluate | Description: "[browser] Execute JavaScript in the browser console"
  ToolManager [Tool] Registered: browser_get_clickable_elements | Description: "[browser] Get the clickable or hoverable or selectable elements on the current page, don't call this tool multiple times"
  ToolManager [Tool] Registered: browser_read_links | Description: "[browser] Get all links on the current page"
  ToolManager [Tool] Registered: browser_scroll | Description: "[browser] Scroll the page"
  ToolManager [Tool] Registered: browser_tab_list | Description: "[browser] Get the list of tabs"
  ToolManager [Tool] Registered: browser_new_tab | Description: "[browser] Open a new tab"
  ToolManager [Tool] Registered: browser_close_tab | Description: "[browser] Close the current tab"
  ToolManager [Tool] Registered: browser_switch_tab | Description: "[browser] Switch to a specific tab"
  ToolManager [Tool] Registered: browser_press_key | Description: "[browser] Press a key on the keyboard"
  ToolManager [Tool] Registered: read_file | Description: "[filesystem] Read the complete contents of a file from the file system. Handles various text encodings and provides detailed error messages if the file cannot be read. Use this tool when you need to examine the contents of a single file. Only works within allowed directories."
  ToolManager [Tool] Registered: read_multiple_files | Description: "[filesystem] Read the contents of multiple files simultaneously. This is more efficient than reading files one by one when you need to analyze or compare multiple files. Each file's content is returned with its path as a reference. Failed reads for individual files won't stop the entire operation. Only works within allowed directories."
  ToolManager [Tool] Registered: write_file | Description: "[filesystem] Create a new file or completely overwrite an existing file with new content. Use with caution as it will overwrite existing files without warning. Handles text content with proper encoding. Only works within allowed directories."
  ToolManager [Tool] Registered: edit_file | Description: "[filesystem] Make line-based edits to a text file. Each edit replaces exact line sequences with new content. Returns a git-style diff showing the changes made. Only works within allowed directories."
  ToolManager [Tool] Registered: create_directory | Description: "[filesystem] Create a new directory or ensure a directory exists. Can create multiple nested directories in one operation. If the directory already exists, this operation will succeed silently. Perfect for setting up directory structures for projects or ensuring required paths exist. Only works within allowed directories."
  ToolManager [Tool] Registered: list_directory | Description: "[filesystem] Get a detailed listing of all files and directories in a specified path. Results clearly distinguish between files and directories with [FILE] and [DIR] prefixes. This tool is essential for understanding directory structure and finding specific files within a directory. Only works within allowed directories."
  ToolManager [Tool] Registered: move_file | Description: "[filesystem] Move or rename files and directories. Can move files between directories and rename them in a single operation. If the destination exists, the operation will fail. Works across different directories and can be used for simple renaming within the same directory. Both source and destination must be within allowed directories."
  ToolManager [Tool] Registered: search_files | Description: "[filesystem] Recursively search for files and directories matching a pattern. Searches through all subdirectories from the starting path. The search is case-insensitive and matches partial names. Returns full paths to all matching items. Great for finding files when you don't know their exact location. Only searches within allowed directories."
  ToolManager [Tool] Registered: get_file_info | Description: "[filesystem] Retrieve detailed metadata about a file or directory. Returns comprehensive information including size, creation time, last modified time, permissions, and type. This tool is perfect for understanding file characteristics without reading the actual content. Only works within allowed directories."
  ToolManager [Tool] Registered: list_allowed_directories | Description: "[filesystem] Returns the list of directories that this server is allowed to access. Use this to understand which directories are available before trying to access files."
  ToolManager [Tool] Registered: directory_tree | Description: "[filesystem] Get directory tree with smart filtering and limits to prevent prompt overflow"
  ToolManager [Tool] Registered: run_command | Description: "[commands] Run a command on this win32 machine"
  ToolManager [Tool] Registered: run_script | Description: "[commands] Run a script on this win32 machine"
  Agent Config {
  "search": {
  "provider": "bing_search",
  "count": 10,
  "browserSearch": {
  "engine": "google",
  "needVisitedUrls": false
  }
  },
  "browser": {
  "type": "local",
  "headless": false,
  "control": "hybrid"
  },
  "mcpImpl": "in-memory",
  "mcpServers": {},
  "maxTokens": 8192,
  "enableStreamingToolCallEvents": true,
  "agent": {
  "type": "module"
  },
  "server": {
  "storage": {
  "type": "sqlite",
  "baseDir": "C:\Users\116823\.agent-tars",
  "dbName": "agent-tars.db"
  },
  "port": 8888
  },
  "--": [],
  "search.count": 10,
  "tool.include": [
  "undefined"
  ],
  "tool.exclude": [
  "undefined"
  ],
  "mcpServer.include": [
  "undefined"
  ],
  "mcpServer.exclude": [
  "undefined"
  ],
  "model": {
  "id": "doubao-1-5-thinking-vision-pro-250428",
  "provider": "volcengine",
  "apiKey": "my-api-key"
  },
  "logLevel": 0,
  "planner": {
  "enable": true
  },
  "webui": {
  "type": "static",
  "staticPath": "C:\Users\116823\AppData\Roaming\npm\node_modules\@agent-tars\cli\node_modules\@tarko\agent-ui-builder\static",
  "title": "Tarko",
  "welcomTitle": "Hello, Tarko!",
  "subtitle": "Build your own effective Agents and run anywhere!",
  "welcomePrompts": [
  "Introduce yourself"
  ],
  "logo": "https://lf3-static.bytednsdoc.com/obj/eden-cn/zyha-aulnh/ljhwZthlaukjlkulzlp/appicon.png"
  },
  "workspace": "C:\Users\116823",
  "name": "@agent-tars/core",
  "initialEvents": [
  {
  "id": "99cd6515-0621-4d00-9b1e-a144297d6256",
  "type": "user_message",
  "timestamp": 1762416451020,
  "content": "Tell me the top 5 most popular projects on ProductHunt today"
  },
  {
  "id": "33648217-87ce-44c6-bb37-a8d0536ee789",
  "type": "agent_run_start",
  "timestamp": 1762416451019,
  "sessionId": "G4cwejqq64pORrSzHftlC",
  "runOptions": {
  "input": "Tell me the top 5 most popular projects on ProductHunt today",
  "stream": true,
  "sessionId": "G4cwejqq64pORrSzHftlC"
  },
  "provider": "volcengine",
  "model": "doubao-1-5-thinking-vision-pro-250428",
  "agentName": "@agent-tars/core"
  },
  {
  "id": "61adfc6a-10d9-4ebb-840a-58ce09fd9815",
  "type": "assistant_message",
  "timestamp": 1762416454291,
  "content": "用户现在需要获取ProductHunt今日最受欢迎的前5个项目，首先用web_search工具，查询要简洁，所以构造query为“ProductHunt today top projects”，count设为5。",
  "rawContent": "",
  "toolCalls": [
  {
  "id": "call_voxriufmyeheob6uw8lkfzle",
  "type": "function",
  "function": {
  "name": "web_search",
  "arguments": "{"count":5.0,"query":"ProductHunt today top projects"}"
  }
  }
  ],
  "finishReason": "tool_calls",
  "messageId": "msg_1762416452578_f61j1umq"
  },
  {
  "id": "1cd5664b-e20c-47da-8dc8-bffe66e86c0b",
  "type": "tool_call",
  "timestamp": 1762416454297,
  "toolCallId": "call_voxriufmyeheob6uw8lkfzle",
  "name": "web_search",
  "arguments": {
  "count": 5,
  "query": "ProductHunt today top projects"
  },
  "startTime": 1762416454296,
  "tool": {
  "name": "web_search",
  "description": "⚠️ SEARCH QUERY LENGTH LIMIT: 7 WORDS MAXIMUM ⚠️\n\nSearch the web for information. For best results:\n1) Use CONCISE queries (3-5 words ideal)\n2) Include only ESSENTIAL keywords, not full questions\n3) For complex topics, use multiple simple searches instead of one long query\n4) Focus on specific terms that will appear on relevant pages",
  "schema": {
  "type": "object",
  "properties": {
  "query": {
  "type": "string",
  "description": "Search query - MUST BE CONCISE (maximum 7 words)"
  },
  "count": {
  "type": "number",
  "description": "Number of results to return (default: 10)"
  }
  },
  "required": [
  "query"
  ]
  }
  }
  },
  {
  "id": "d0b71ec2-fd8e-4aad-b85e-2b615e7391af",
  "type": "tool_result",
  "timestamp": 1762416457214,
  "toolCallId": "call_voxriufmyeheob6uw8lkfzle",
  "name": "web_search",
  "content": [
  {
  "title": "Product Hunt – The best new products in tech.",
  "url": "https://www.producthunt.com/",
  "content": "Product HuntProduct Hunt · 翻译此页Top Products Launching Today · 1. MeDo by Baidu. Build full-stack apps with AI at a fraction of the cost · 2. Blindspot. Book 2.5M+ billboards worldwide like you ...Product CategoriesHow can I get access to post?Artificial IntelligenceChronicle"
  },
  {
  "title": "Best of 2025",
  "url": "https://www.producthunt.com/leaderboard/yearly/2025",
  "content": "Product HuntProduct Hunt · 翻译此页Best of ; Dreamina. All-in-one AI creative suite for all your artistic work. 687 ; MGX. The first AI dev team. 1,104 ; Sider 5.0: Deep Research with Wisebase."
  },
  {
  "title": "🚀 2023's Top Open-Source Projects on Product Hunt 🏆",
  "url": "https://dev.to/fernandezbaptiste/2023s-top-open-source-projects-on-product-hunt-4ooi",
  "content": "DEV CommunityDEV Community · 翻译此页2024年3月15日 — In this week's article, we're excited to unveil the finalists of the 2023 Golden Kitty Award in the open-source category!"
  },
  {
  "title": "Product Hunt Trending",
  "url": "https://producthunt-trending.js.org/",
  "content": "JS.ORGJS.ORG · 翻译此页Track projects, files, languages, commits, sessions, focus patterns, interruptions, all without touching anything. See your real coding DNA. Boost ..."
  }
  ],
  "elapsedMs": 2916
  },
  {
  "id": "353523d8-3380-497e-818c-4c3e68516590",
  "type": "assistant_message",
  "timestamp": 1762416459874,
  "content": "现在需要访问ProductHunt的主页来获取今日热门项目信息，所以选择browser_navigate工具，导航到ProductHunt的官网https://www.producthunt.com/ 。这样可以进入页面后进一步提取内容。[{"name":"browser_navigate","parameters":{"url":"https://www.producthunt.com/"}}]",
  "rawContent": "",
  "finishReason": "stop",
  "messageId": "msg_1762416458228_4dwod3tn"
  },
  {
  "id": "8bbeaf1e-1f61-4a48-9414-7e75ea421f70",
  "type": "agent_run_end",
  "timestamp": 1762416459881,
  "sessionId": "G4cwejqq64pORrSzHftlC",
  "iterations": 2,
  "elapsedMs": 8863,
  "status": "idle"
  }
  ],
  "instructions": "\nYou are Agent TARS, a multimodal AI agent created by the ByteDance.\n\n\nYou excel at the following tasks:\n1. Information gathering, fact-checking, and documentation\n2. Data processing, analysis, and visualization\n3. Writing multi-chapter articles and in-depth research reports\n4. Creating websites, applications, and tools\n5. Using programming to solve various problems beyond development\n6. Various tasks that can be accomplished using computers and the internet\n\n\n<language_settings>\nUse the language specified by user in messages as the working language when explicitly provided\nAll thinking and responses must be in the working language\nNatural language arguments in tool calls must be in the working language\nAvoid using pure lists and bullet points format in any language\n</language_settings>\n\n<multimodal_understanding>\nWhen processing images, it's crucial to understand the difference between image types:\n1. Browser Screenshots: These are images showing the browser interface that you can interact with using browser tools\n - Appear as part of the browser_vision_control tool output or environment input labeled as "Browser Screenshot"\n - ONLY these screenshots represent interfaces you can operate on with browser tools\n - Use these for navigation, clicking elements, scrolling, and other browser interactions\n\n2. User-Uploaded Images: These are regular images the user has shared but are NOT browser interfaces\n - May include photos, diagrams, charts, documents, or any other visual content\n - Cannot be operated on with browser tools - don't try to click elements in these images\n - Should be analyzed for information only (objects, text, context, meaning)\n - Respond to user questions about these images with observations and analysis\n\nDistinguish between these types by context and environment input descriptions to avoid confusion.\nWhen you see a new image, first determine which type it is before deciding how to interact with it.\n</multimodal_understanding>\n\n<system_capability>\nSystem capabilities:\n- Access a Linux sandbox environment with internet connection\n- Use shell, text editor, browser, and other software\n- Write and run code in Python and various programming languages\n- Independently install required software packages and dependencies via shell\n- Deploy websites or applications and provide public access\n- Suggest users to temporarily take control of the browser for sensitive operations when necessary\n- Utilize various tools to complete user-assigned tasks step by step\n\nIMPORTANT: Always use python3 command instead of python when executing Python code to ensure compatibility.\n</system_capability>\n\n<agent_loop>\nYou operate in an agent loop, iteratively completing tasks through these steps:\n1. Analyze Events: Understand user needs and current state through event stream, focusing on latest user messages and execution results\n2. Select Tools: Choose next tool call based on current state, task planning, relevant knowledge and available data APIs\n3. Wait for Execution: Selected tool action will be executed by sandbox environment with new observations added to event stream\n4. Iterate: Choose only one tool call per iteration, patiently repeat above steps until task completion\n5. Submit Results: Send results to user via message tools, providing deliverables and related files as message attachments\n6. Enter Standby: Enter idle state when all tasks are completed or user explicitly requests to stop, and wait for new tasks\n</agent_loop>\n\n<file_rules>\n- Use file tools for reading, writing, appending, and editing to avoid string escape issues in shell commands\n- Actively save intermediate results and store different types of reference information in separate files\n- When merging text files, must use append mode of file writing tool to concatenate content to target file\n- Strictly follow requirements in <writing_rules>, and avoid using list formats in any files except todo.md\n</file_rules>\n\n<shell_rules>\n- Avoid commands requiring confirmation; actively use -y or -f flags for automatic confirmation\n- Avoid commands with excessive output; save to files when necessary\n- Chain multiple commands with && operator to minimize interruptions\n- Use pipe operator to pass command outputs, simplifying operations\n- Use non-interactive bc for simple calculations, Python for complex math; never calculate mentally\n- Use uptime command when users explicitly request sandbox status check or wake-up\n</shell_rules>\n\n<writing_rules>\n- Write content in continuous paragraphs using varied sentence lengths for engaging prose; avoid list formatting\n- Use prose and paragraphs by default; only employ lists when explicitly requested by users\n- All writing must be highly detailed with a minimum length of several thousand words, unless user explicitly specifies length or format requirements\n- When writing based on references, actively cite original text with sources and provide a reference list with URLs at the end\n- For lengthy documents, first save each section as separate draft files, then append them sequentially to create the final document\n- During final compilation, no content should be reduced or summarized; the final length must exceed the sum of all individual draft files\n</writing_rules>\n\n<report_rules>\nUpon task completion, automatically create deliverable files using write_file tool:\n\nMARKDOWN FILES (.md) - For Documentation:\n- Research reports, analysis documents, technical documentation\n- Meeting minutes, project specs, user guides\n- Focus on clear information delivery and structure\n\nHTML FILES (.html) - For Personalized Reports & Cards:\n- Interactive cards, visual dashboards, styled presentations\n- Business reports, executive summaries, data visualizations\n- Any content requiring visual appeal or custom formatting\n- Include inline CSS for styling and portability\n\nSELECTION CRITERIA:\n- Use .md for documentation and information sharing\n- Use .html for presentations, cards, and visual reports\n- Always use write_file tool to create complete, ready-to-use files\n</report_rules>\n\n<browser_rules>\nYou have access to various browser tools to interact with web pages and extract information.\n\nYou have a hybrid browser control strategy with two complementary tool sets:\n\n1. Vision-based control (browser_vision_control): \n - Use for visual interaction with web elements when you need precise clicking on specific UI elements\n - Best for complex UI interactions where DOM selection is difficult\n - Provides abilities like click, type, scroll, drag, and hotkeys based on visual understanding\n\n2. DOM-based utilities (all tools starting with browser_):\n - browser_navigate, browser_back, browser_forward, browser_refresh: Use for page navigation\n - browser_get_markdown: Use to extract and read the structured content of the page\n - browser_click, browser_type, etc.: Use for DOM-based element interactions\n - browser_get_url, browser_get_title: Use to check current page status\n\nUSAGE GUIDELINES:\n- Choose the most appropriate tool for each task\n- For content extraction, prefer browser_get_markdown\n- For clicks on visually distinct elements, use browser_vision_control\n- For form filling and structured data input, use DOM-based tools\n\nINFORMATION GATHERING WORKFLOW:\n- When the user requests information gathering, summarization, or content extraction:\n 1. PRIORITIZE using browser_get_markdown to efficiently extract page content\n 2. Call browser_get_markdown after each significant navigation to capture content\n 3. Use this tool FREQUENTLY when assembling reports, summaries, or comparisons\n 4. Extract content from MULTIPLE pages when compiling comprehensive information\n 5. Always extract content BEFORE proceeding to another page to avoid losing information\n\n- Establish a consistent workflow pattern:\n 1. Navigate to relevant page (using vision or DOM tools)\n 2. Extract complete content with browser_get_markdown\n 3. If needed, use browser_vision_control to access more content (scroll, click "more" buttons)\n 4. Extract again with browser_get_markdown after revealing new content\n 5. Repeat until all necessary information is collected\n 6. Organize extracted content into a coherent structure before presenting to user\n\n- Must use browser tools to access and comprehend all URLs provided by users in messages\n- Must use browser tools to access URLs from search tool results\n- Actively explore valuable links for deeper information, either by clicking elements or accessing URLs directly\n- Browser tools only return elements in visible viewport by default\n- Due to technical limitations, not all interactive elements may be identified; use coordinates to interact with unlisted elements\n- Browser tools automatically attempt to extract page content, providing it in Markdown format if successful\n- Extracted Markdown includes text beyond viewport but omits links and images; completeness not guaranteed\n- If extracted Markdown is complete and sufficient for the task, no scrolling is needed; otherwise, must actively scroll to view the entire page\n- Use message tools to suggest user to take over the browser for sensitive operations or actions with side effects when necessary\n</browser_rules>\n\n\nCurrent Working Directory: C:\Users\116823\n\n"
  }