The Agent Skills Directory

[PROMPT_INJECTION]: No evidence of prompt injection or instructions to bypass safety guidelines was found. The skill uses standard instructional constraints to limit knowledge to official documentation.
[DATA_EXFILTRATION]: No hardcoded credentials, sensitive file path access, or unauthorized network operations were detected. The only external URLs are to the official Sui documentation and a well-known MCP documentation server.
[REMOTE_CODE_EXECUTION]: There are no patterns for remote script execution (e.g., curl|bash) or unverifiable package installations.
[OBFUSCATION]: The skill content and code samples were checked for Base64 encoding, zero-width characters, homoglyphs, and other hidden content; no such patterns were found.
[PRIVILEGE_ESCALATION]: The skill does not contain any commands associated with acquiring higher system permissions or modifying system configurations.
[DYNAMIC_EXECUTION]: The Move code samples provided are for reference only and do not involve runtime code generation, unsafe deserialization, or self-modifying logic.
[INDIRECT_PROMPT_INJECTION]: The skill is a static reference for Move development and does not appear to have an ingestion surface for untrusted external data that would influence the agent's behavior.

kiosk