llm-monitoring-dashboard
Fail
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: CRITICALREMOTE_CODE_EXECUTIONPERSISTENCEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill directs the execution of a remote shell script using the
curl | bashpattern from an untrusted third-party GitHub repository (nooscraft/tokuin). This method allows unverified remote code to be executed with the same privileges as the user, as seen in Step 1 and the generated safety-guard.sh script. - [PERSISTENCE]: The skill creates multiple persistent background processes by modifying the system crontab. It schedules scripts for metrics collection (every 5 minutes), report generation (weekly), and cost alerts (hourly), ensuring the skill's logic continues to run automatically across user sessions.
- [EXTERNAL_DOWNLOADS]: The skill downloads content from
raw.githubusercontent.com/nooscraft/tokuin, which is not a recognized trusted organization or well-known service. It also performs external package installations via NPM for components such as recharts and better-sqlite3. - [COMMAND_EXECUTION]: The skill generates and executes several local shell and Python scripts. It also starts a local web server using Python's
http.servermodule, which exposes a local port for monitoring access.
Recommendations
- HIGH: Downloads and executes remote code from: https://raw.githubusercontent.com/nooscraft/tokuin/main/install.sh - DO NOT USE without thorough review
- AI detected serious security threats
Audit Metadata