fdk-uat
Warn
Audited by Snyk on Jul 21, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.85). SKILL.md yêu cầu runtime workflow chạy
curl -fsSL "$RAW/.../bootstrap.sh" | bashvàcurl ... "$RAW/.../policy.yaml" | grep ..., nơi$RAWtrỏ tới raw content của nhánh/canary trên GitHub (outsider-authored public web text), rồi nội dung đó được đưa vào ngữ cảnh LLM thông qua việc thực thi script/harness.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 1.00). The skill fetches and executes remote install scripts at runtime (e.g., via curl -fsSL "$RAW/harness/poc-vendor-neutral/bootstrap.sh" | bash) where RAW is set to https://raw.githubusercontent.com///$CANARY (and similarly the main-URL https://raw.githubusercontent.com///<nhánh chính>), so remote content directly executes code and controls the install flow.
Issues (2)
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
Audit Metadata