Open-source AI browser agent for Chrome and Firefox (monorepo) π§
JavaScript ai-agentai-agentsai-toolsbrowser-aibrowser-automationbrowser-extensionchrome-aichrome-devtoolschrome-extensionchrome-extensionsfirefox-addonfirefox-addonsfirefox-extensionfirefox-extensionsgemmagemma4minimaxoffline-aioffline-llmqwen
{
"id": "011",
"category": "tool-error-pivot",
"mode": "act",
"browser": "chrome",
"tab": {
"url": "https://mastoturk.org/publish",
"title": "MastoTurk"
},
"description": "inspect_viewport returned \"no vision capability\". Must pivot to AX tree, not retry the capture.",
"seed": [
{
"role": "user",
"content": "[Current page context β applies to this user message and supersedes older page context for phrases like \"this page\". URL: https://mastoturk.org/publish β Title: MastoTurk]\n\nhow many characters does the post have so far?"
},
{
"role": "assistant",
"content": "",
"tool_calls": [
{
"id": "call_e_17eb23",
"type": "function",
"function": {
"name": "inspect_viewport",
"arguments": "{}"
}
}
]
},
{
"role": "tool",
"tool_call_id": "call_e_17eb23",
"name": "inspect_viewport",
"content": "{\"success\":false,\"error\":\"This model cannot see images: it has no vision capability and no dedicated vision model is configured. In provider settings, enable \\\"Model supports vision\\\" for the active provider or set a vision model. For now, use get_accessibility_tree, get_interactive_elements, or read_page to inspect the page.\"}"
}
],
"expected": {
"idealNextToolCall": {
"name": "get_accessibility_tree",
"args": {
"filter": "visible"
}
},
"antiPatterns": [
{
"match": "inspect_viewport({})",
"reason": "Retrying a tool that just declared it cannot work β gpt-4o trace xpx9t9 made this mistake."
}
],
"successRubric": "Pivot to a text-based inspection tool (get_accessibility_tree, read_page, or get_interactive_elements). The error message even names the alternatives β use one of them."
}
}