analyze-traj
Analyze OSWorld-V2 agent trajectory logs and task results to produce actionable insights. Use this skill whenever the user wants to understand agent performance on OSWorld tasks — including analyzing trajectories, reviewing task results, finding error patterns, comparing code vs GUI strategies, identifying which tools/commands the agent used, or deciding which task types to scale up in the benchmark.
git clone --depth 1 https://github.com/AMAP-ML/LongHorizon-Harness /tmp/analyze-traj && cp -r /tmp/analyze-traj/eval/OSWorldv2-harness/OSWorld-V2/.codex/skills/analyze-traj ~/.claude/skills/analyze-trajSKILL.md
If only one task is issued, analyze it directly with instruction: [analyze-single-traj.md](prompts/analyze-single-traj.md). If multiple tasks or a whole results directory are issued, use subagents to analyze them in parallel (one agent for each task). Do not analyze them sequentially by yourself. DO NOT tell it what to do. Just ask the subagent to analyze the task in target directory and use this skill (`analyze-traj`) to do the analysis. Pass any user instructions to every subagent. After the per-task reports are ready: - Do nothing but report to the user that the analysis is done and where to find the reports. - Ask user if they want to synthesize a run-level summary, if yes use: [analyze-full-run.md](prompts/analyze-full-run.md).
Check OSWorld tasks. Validate the evaluation function, verify that the instruction is feasible given the task setup and agent-visible files, inspect setup artifacts when needed, and produce both markdown and structured JSON reports.
Migrate an agent from upstream OSWorld into this OSWorld-V2 repository, add matching evaluation entrypoints, and verify the integration.
Provision and verify an OSWorld-V2 checkout after clone. Use when the user asks for OSWorld-V2 setup, installation, onboarding, AWS provider setup, Docker provider setup, mocked website server setup, GitLab server setup, gated task download, CUA-Harness hybrid experiment setup, or a final runnable export block. The skill should install/configure the selected supported infrastructure where possible, ask for user confirmation or credentials when required, and report what is fully configured versus still blocked.
Reproduce CUA-Harness experiments on WeaveBench from a GitHub checkout. Use when the user wants an AI coding agent to set up dependencies, download WeaveBench assets, prepare the 120G VM, configure Qwen/Anthropic-compatible APIs, run smoke tests, launch full or subset evaluations, inspect logs, or summarize scores for this repository.