codex-claude-cursor-loop
Fail
Audited by Snyk on Aug 24, 2026
Risk Level: HIGH
Full Analysis
HIGH W007: Insecure credential handling detected in skill instructions.
- Insecure credential handling detected (high risk: 0.80). The skill repeatedly instructs storing a Cursor session ID and embedding it verbatim in command-line --resume arguments (e.g., cursor-agent --resume=""), which requires the LLM to handle and reproduce a secret-like token in its outputs, creating an exfiltration risk.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (medium risk: 0.30). 이 스킬은 런타임에
AskUserQuestion/사용자 입력으로 받은 “계획”과 “구현 설명/코드(파일 변경 요약 포함)” 같은 텍스트를 Codex/Cursor 평가 프롬프트에 직접 포함시키므로, 제3자가 자유 텍스트를 POST하든(예: 사용자 질의로) 그 내용을 LLM이 선택 없이 즉시 읽게 됩니다.
Issues (2)
W007
HIGHInsecure credential handling detected in skill instructions.
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
Audit Metadata