advanced-keigo-usage-expert

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns detected. The skill consists of instructional Markdown files without executable code, scripts, or dangerous commands.
  • [PROMPT_INJECTION]: The instructions focus on domain expertise in Japanese Keigo and do not contain attempts to bypass AI safety guidelines or override system instructions.
  • [DATA_EXFILTRATION]: No access to sensitive files or network calls to exfiltrate data were found.
  • [REMOTE_CODE_EXECUTION]: The skill does not download or execute any remote scripts or packages.
  • [INDIRECT_PROMPT_INJECTION]: The skill suggests using platform tools to ingest data from external sources (Wikipedia, ArXiv) for research purposes.
  • Ingestion points: modules/research-checklist.md recommends using research tools to fetch data from Wikipedia and ArXiv.
  • Boundary markers: No explicit boundary markers or "ignore instructions" wrappers are defined for the ingested data.
  • Capability inventory: No dangerous capabilities (file writing, shell execution, or network exfiltration) are present in the provided skill files.
  • Sanitization: No sanitization or filtering of external content is specified.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 05:38 PM
Security Audit — agent-trust-hub — advanced-keigo-usage-expert