Repository Issue Activity (beta)

#195 [Bug] WindowsACI.hotkey() splits key names into individual characters

#201 Arbitrary code execution: model-grounded action string passed to eval() in create_pyautogui_code

#199 Prompt Injection in Agent-S can reach `exec(code[0])` and trigger unintended file write

#196 Permission dialogs execute unescaped model-generated text via os.system on macOS/Linux

#44 Ollama support?

#178 Reduce grounding model call frequency

#189 Feature: Governance controls for computer-using agents

#187 Sidebar broken Windows 11

#186 Code execution tool broken

#185 Conversation with AI flat out broken

#184 Website security audit findings — simular.ai

#181 Is this still alive? I asked Claude to analyze this repo and it shocked me.

#183 [AI Village Outreach] Hello from the AI Village agents!

#182 Feature: AgentID for computer-use agent authentication

#175 The problem of coordinate offset

#167 Problem with bBoN

#176 GLM-4.7 and GLM-Image have any demo?

#172 Fixed interval schedules

#171 Fire Kirin

#154 What is the difference between s2.5 and s2?

#152 time to run in OSWorld with AgentS3

#149 Task execution time (latency) of Agent-S

#168 Attempt 1 failed: Error code: 400 - {'error': {'message': 'At most 3 image(s) may be provided in one prompt. None', 'type': 'BadRequestError', 'param': None, 'code': 400}}

#61 What knowledge / memory does the reported performance on OSWorld need?

#158 Does this framework have plans to support Qwen and DeepSeek?