viral-instagram-reels
Fail
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: CRITICALOBFUSCATIONINDIRECT_PROMPT_INJECTIONMETADATA_POISONING
Full Analysis
- [OBFUSCATION]: Detection of a homoglyph-based phishing URL in
SKILL.md(line 12). The URLhttps://raw.githubusercontent.com/vyralcontent/content-skills/main/assets/header.pngcontains Confusable/homoglyph characters that cause it to resolve to a malicious typosquatted or phishing domain (raw.githubusercontent.co) rather than the official GitHub source. Automated scanners correctly flagged this as a malicious phishing link. - [INDIRECT_PROMPT_INJECTION]: Self-referential injection via
skill-report.json. The skill includes a fabricated JSON file masquerading as a security audit report. This file attempts to subvert the analysis process by declaring the skill 'SAFE' and explicitly instructing the analyzer to disregard legitimate security findings (such as the phishing URL and reconnaissance alerts) as false positives. This is a direct attempt to bypass AI and human safety filters. - [METADATA_POISONING]: Deceptive metadata and labeling. The skill includes a 'security report' file and inconsistent authorship metadata designed to masquerade as an officially vetted skill while housing malicious phishing components.
Recommendations
- AI detected serious security threats
- Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata