utility-tools

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • PROMPT_INJECTION (LOW): The skill is susceptible to indirect prompt injection through tools that ingest untrusted data.
  • Ingestion points: The /v2/og endpoint fetches metadata from external URLs, and the /v2/fanyi endpoint processes arbitrary user text.
  • Boundary markers: Absent. There are no instructions to the agent to ignore or sanitize embedded commands within the processed data.
  • Capability inventory: The skill utilizes the requests library for network operations and performs file writes (e.g., qrcode.png).
  • Sanitization: No evidence of sanitization or content validation for the API responses before they are returned to the agent context.
  • EXTERNAL_DOWNLOADS (LOW): The skill relies on an external API (60s.viki.moe) that is not part of the trusted source whitelist.
  • It downloads and saves binary content (QR code images) to the local filesystem.
  • All core functionality is dependent on this third-party service, which could be modified to return malicious payloads or instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:36 PM