NYC

devops-flow

Fail

Audited by Snyk on Feb 15, 2026

Risk Level: HIGH
Full Analysis

HIGH W007: Insecure credential handling detected in skill instructions.

  • Insecure credential handling detected (high risk: 0.90). The prompt includes generated templates that embed plaintext credentials (e.g., docker-compose DATABASE_URL and POSTGRES_PASSWORD) and scripts that load/export .env files, which can require the LLM to output secret values verbatim and thus pose secret-exfiltration risk.

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

  • Attempt to modify system services in skill instructions detected (low risk: 0.30). The skill generates and encourages execution of scripts and Ansible playbooks that perform privileged operations (become: yes, systemd restarts, file ownership changes, installs and command execution) and would modify the machine or services if targeted at the local host, but it does not explicitly instruct obtaining/bypassing sudo or creating host user accounts outside container images.
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 15, 2026, 09:16 PM