skill-creator

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The repository includes a deliberate "dynamic shell injection" feature in SKILL.md that executes embedded shell commands (RCE risk), plus code that force-kills processes and loads remote CDN resources — together these enable remote/host compromise or data exposure if abused.

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

• Third-party content exposure detected (medium risk: 0.30). In the required eval/review workflow, the agent reads user-authored free-text feedback from feedback.json (posted via the eval viewer) and also reads the executed run artifacts/transcripts (including prompts/outputs) from the workspace for grading and later analysis.