nagios-expert

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

• Third-party content exposure detected (high risk: 0.70). The skill's core references include instructions and examples that fetch and act on arbitrary public web content (e.g., references/programming.md's check_webapp_health.py which uses urllib to GET an arbitrary URL and decide CRITICAL/OK, references/plugins.md and check_http examples, and installation.md/wget instructions to pull code from GitHub), so the agent/workflow is clearly described to ingest untrusted third‑party content that can change behavior.

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

• Attempt to modify system services in skill instructions detected (high risk: 0.90). This Nagios setup/configuration skill explicitly covers installation, users/groups, Apache/httpd and service startup, NRPE and system-level configuration—tasks that typically require creating accounts and editing system/service files with sudo and thus can push an agent to modify the host state.