skills/mystenlabs/skills/zklogin/Gen Agent Trust Hub

zklogin

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references and retrieves technical information from official Sui documentation and vendor-controlled infrastructure.\n
  • Evidence: Information is sourced from docs.sui.io and utilizes the mystenlabs.com GraphQL endpoint for offchain verification (integration.md).\n
  • Context: These sources are trusted and official components of the Sui ecosystem managed by the skill's author.\n- [INDIRECT_PROMPT_INJECTION]: The skill uses an external MCP server for documentation lookups, which constitutes an ingestion point for external data.\n
  • Ingestion points: External documentation tool at https://sui.mcp.kapa.ai (SKILL.md).\n
  • Boundary markers: Not present; the instructions do not provide specific delimiters to separate the tool's response from other system instructions.\n
  • Capability inventory: The skill reference files (architecture.md, integration.md) do not contain scripts or instructions for performing high-risk operations like shell command execution or direct network exfiltration.\n
  • Sanitization: No sanitization or validation logic is defined for the external tool output, although the source is a reputable documentation provider.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 03:15 PM
Security Audit — agent-trust-hub — zklogin