a365-setup

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly includes a command that prints a stored client secret ("a365 setup blueprint --show-secret") and also mandates showing CLI output verbatim, which forces the LLM to display secret values if those commands are run.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 1.00). These URLs point to direct PowerShell/bash installer scripts hosted on a third-party domain (astral.sh) and are referenced for pipe-to-shell installation, which is a high-risk pattern for distributing malware or executing untrusted code.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill includes explicit runtime install commands that fetch and execute remote scripts (curl | sudo bash and irm | iex / curl | sh), e.g. https://aka.ms/InstallAzureCLIDeb and https://astral.sh/uv/install.sh (and https://astral.sh/uv/install.ps1), which are runtime external dependencies that execute remote code.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.90). The skill explicitly instructs running system package installation and setup commands that require sudo (e.g., apt installs and piping a script to sudo), which directs the agent to modify the host system state and escalate privileges.