MyPCBench: A Benchmark for Personally Intelligent Computer-Use Agents
This benchmark addresses the gap in evaluating personal assistants that must handle logged-in accounts and personal context, providing a more realistic test for computer-use agents.
MyPCBench introduces a benchmark for personal computer-use agents using a simulated Linux desktop with 17 web applications and 184 tasks based on real user requests. The best model, Claude Opus 4.6, solves 55.4% of tasks, with failures concentrated on multi-application and long-horizon tasks.
Current benchmarks for computer-use agents evaluate models in impersonal environments. This leaves a gap between evaluation and deployment where personal assistants are expected to work across a user's whole digital life, including their context, historical data, and logged-in accounts. This gap is widest on web tasks, where live web evaluations cannot exercise sites that require logging in or personal information, the kind of site a real personal assistant has to drive. We introduce MyPCBench, which tests computer-use agents as personal assistants on a Linux desktop populated with 17 simulated real-world web applications and a full desktop stack, all seeded for one canonical persona, Michael Scott from The Office. We define 184 tasks in this environment, each inspired by a real request drawn from the OpenClaw community, and benchmark six closed and open-weight models with a uniform computer+bash tool surface. We find that the best model, Claude Opus 4.6, fully solves 55.4\% of the tasks, the only model above 50\%. Model failures cluster on tasks that span many applications and on long trajectories, where personalization stresses an assistant the most. We release the environment, task set, and agent harness at https://mypcbench.com.