llm-monitoring-dashboard

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (medium risk: 0.65). Most links point to reputable CDNs and known services (Google Fonts, jsDelivr, Chart.js, OpenRouter), but the presence of raw install scripts (raw.githubusercontent.com/*.sh and *.ps1) and curl|bash / iwr|iex install instructions from an unverified GitHub user (nooscraft) is a high‑risk pattern for malware distribution if the repo or account is not trusted.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill's required Step 1 installation commands explicitly fetch and execute a public install script (curl -fsSL https://raw.githubusercontent.com/nooscraft/tokuin/main/install.sh | bash) from raw.githubusercontent.com, which is untrusted third‑party content that would be run/interpreted as part of the workflow and could therefore inject instructions affecting tool behavior.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly runs remote installer commands (curl -fsSL https://raw.githubusercontent.com/nooscraft/tokuin/main/install.sh | bash and the PowerShell equivalent https://raw.githubusercontent.com/nooscraft/tokuin/main/install.ps1 | iex) at runtime to fetch and execute the Tokuin install script, which is a required dependency and executes remote code.