omd-harness

Warn

Audited by Snyk on Aug 26, 2026

Risk Level: MEDIUM
Full Analysis

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

  • Third-party content exposure detected (medium risk: 0.30). SKILL.md의 런타임 경로는 Step 2.5/3/4에서 사용자 자연어 task와 채널 데이터의 로컬 파일들(예: ctx-prime.json, reference-fingerprints.json, reference-tags.md, vocabulary.json, references//DESIGN.md) 및 캡처 산출물을 LLM에 전달/읽는 흐름이며, outsider가 자유 텍스트를 사용자 입력(“/omd-harness ”) 형태로 직접 주입해 master의 프롬프트/체크포인트에 영향을 줄 수 있다.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

  • Potentially malicious external URL detected (high risk: 0.70). The skill explicitly lists "https://oh-my-design.kr//design.md" as a runtime catalog resolution fallback for resolving reference DESIGN.md, which would be fetched at runtime and can directly influence master prompts/instructions.

Issues (2)

W011
MEDIUM

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

W012
MEDIUM

Unverifiable external dependency detected (runtime URL that controls agent).

Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 26, 2026, 08:34 AM
Issues
2
Security Audit — snyk — omd-harness