performing-endpoint-vulnerability-remediation

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

• Third-party content exposure detected (high risk: 0.70). The skill's scripts (scripts/process.py) explicitly fetch and ingest public third‑party feeds — e.g., the CISA KEV JSON at https://www.cisa.gov/sites/default/files/feeds/known_exploited_vulnerabilities.json and the EPSS API at https://api.first.org/data/v1/epss — and those external values are used to prioritize vulnerabilities and set SLA/ remediation decisions, so external content can materially influence agent behavior.

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill includes explicit privileged system-modifying instructions (PowerShell commands to install updates, change services, and edit HKLM registry keys) that direct the agent to alter the machine's state and require administrative/sudo privileges.