legal-contract-review
Warn
Audited by Snyk on Jun 17, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill explicitly performs runtime WebFetches (cache-then-fetch) to verify statutes and cases—e.g. https://law.moj.gov.tw and https://judgment.judicial.gov.tw per assets/legal-sources.json—and the fetched content is injected into the LLM verification/comparison pipeline (controlling citation verification and LLM prompts), so these runtime URLs are external dependencies that directly influence agent prompts/outputs.
MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).
- Hidden Unicode characters detected (1 type(s) found)
Issues (2)
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
W021
MEDIUMHidden or invisible Unicode characters detected (potential obfuscation or prompt injection).
Audit Metadata