skills/nakajima/pi-rot/svelte/Gen Agent Trust Hub

svelte

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is composed of architectural guidance and code snippets for Svelte 5. No malicious functionality is present.
  • [PROMPT_INJECTION]: No attempts to override agent instructions or bypass safety guardrails were detected in the guidance.
  • [DATA_EXFILTRATION]: No sensitive file access or unauthorized network exfiltration patterns were found. Documentation examples use generic placeholders.
  • [REMOTE_CODE_EXECUTION]: No remote code execution or untrusted script downloading patterns exist. References to libraries like 'tippy.js' are standard development practices.
  • [COMMAND_EXECUTION]: No use of shell commands, subprocesses, or dynamic execution of untrusted input.
  • [CREDENTIALS_UNSAFE]: No hardcoded secrets, API keys, or private tokens are present in the documentation or examples.
  • [OBFUSCATION]: The content does not use any encoding, hidden characters, or other techniques to mask intent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 04:46 AM
Security Audit — agent-trust-hub — svelte