linux-patcher
Warn
Audited by Snyk on Feb 16, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.80). The skill queries an external PatchMon API (scripts/patchmon-query.sh uses PATCHMON_URL/api/v1/dashboard/hosts) and ingests/parses host data (hostname, sshUser, dockerPath) from that third-party/untrusted source into its runtime workflow, so arbitrary user-generated or external data can influence the agent's actions.
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 requires and instructs configuring passwordless sudo and editing /etc/sudoers.d, and it has the agent run sudo package/docker commands and install/configure system-level tools and cron jobs—actions that modify system files and perform privileged operations on the host(s).
Audit Metadata