computer-use

Warn

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references and installs external components using npx skills add. These components originate from third-party repositories including stablyai/orca@computer-use, web-infra-dev/midscene-skills@computer-automation, and am-will/codex-skills@gemini-computer-use, which are not verified trusted sources.
  • [COMMAND_EXECUTION]: The script scripts/computer_use_runner.py executes shell commands using subprocess.run(shell=True). This is used to programmatically trigger the installation of fallback skills based on a provided index.
  • [REMOTE_CODE_EXECUTION]: The mechanism for adding fallback skills allows for the ingestion of remote code and instructions from unverified third-party registries. This creates a supply chain risk where the agent may execute logic controlled by external entities.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 29, 2026, 03:52 AM
Security Audit — agent-trust-hub — computer-use