claude-codex-usage-dashboard

Fail

Audited by Gen Agent Trust Hub on Jul 14, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Downloads the dashboard codebase from an external GitHub repository: https://github.com/frankchiu-dev/claude-codex-usage-dashboard.git.
  • [REMOTE_CODE_EXECUTION]: Instructs users to execute the downloaded application and associated scripts (node server.js, setup-claude-statusline.sh, start-dashboard.sh) which perform various system-level operations.
  • [COMMAND_EXECUTION]: Performs several high-risk command-line operations including:
  • Modifying Windows Firewall rules using netsh advfirewall firewall add rule to allow incoming traffic.
  • Establishing system persistence using launchctl on macOS to ensure the server starts automatically at login.
  • Executing user-defined shell commands or scripts through the EXTRA_STATUSLINE_COMMAND environment variable or config.json configuration.
  • Granting execution permissions to downloaded scripts via chmod +x.
  • [DATA_EXFILTRATION]: Accesses local configuration and usage files in the user's home directory (~/.claude/usage-cache.json and ~/.codex/sessions) and serves this data over the local network. This makes potentially sensitive API usage information available to any device connected to the same Wi-Fi network.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Jul 14, 2026, 12:29 AM
Security Audit — agent-trust-hub — claude-codex-usage-dashboard