Scale computer-use 2.0 with open-source drivers, cross-OS sandboxes, and benchmarks for training, evaluation, and data generation.
Rust agentai-agentapplecomputer-usecomputer-use-agentcontainerizationcuadesktop-automationhacktoberfestlumemacosmanusoperatorswiftvirtualizationvirtualization-frameworkwindowswindows-sandbox
Added agent tool filtering
Dillon DuPontc58ff5596920
diff --git a/libs/python/agent/agent/integrations/hud/__init__.py b/libs/python/agent/agent/integrations/hud/__init__.pyindex 8a203e0e45..e27060ff03 100644--- a/libs/python/agent/agent/integrations/hud/__init__.py+++ b/libs/python/agent/agent/integrations/hud/__init__.py@@ -11,6 +11,7 @@ Exports:import timefrom typing import Any, Optional+from agent.computers import is_agent_computerfrom datasets import load_dataset, Datasetfrom hud.datasets import Task, run_datasetfrom hud import trace@@ -55,6 +56,15 @@ async def run_single_task(sample_task = dataset[task_id] # type: ignore[index]task_prompt = sample_task.get("prompt", f"Task {sample_task.get('id', 0)}") # type: ignore[attr-defined]+ # Filter any existing Computer tools+ # The eval framework will add its own Computer tool per task+ if tools:+ tools = [+ tool+ for tool in tools+ if not is_agent_computer(tool)+ ]+with trace(name=task_prompt):task = Task(**sample_task) # type: ignore[arg-type]@@ -118,6 +128,15 @@ async def run_full_dataset(dataset_name = "custom"job_name = job_name or f"Evaluation {time.strftime('%H:%M %Y-%m-%d')}"+ # Filter any existing Computer tools+ # The eval framework will add its own Computer tool per task+ if tools:+ tools = [+ tool+ for tool in tools+ if not is_agent_computer(tool)+ ]+# Execute evaluationreturn await run_dataset(name=job_name,