Skills
skills/gasserane/personal-skills/instrument-designer/Gen Agent Trust Hub

instrument-designer

Pass

Audited by Gen Agent Trust Hub on Jul 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: Comprehensive analysis of the instructions and logical flow found no evidence of malicious patterns, obfuscation, or unauthorized system access.\n- [DATA_EXFILTRATION]: The skill uses vendor-provided modules (ane_package.reporting.word_export and xlsform.write_xlsform) to export the designed instruments to human-readable Word files and machine-readable survey forms. These operations are essential to the skill's purpose and utilize authorized vendor infrastructure.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest research targets and questions provided by the user to generate instruments. This functionality represents a standard input-output workflow and does not involve executing the ingested data as code or commands.\n
  • Ingestion points: Measurement target and Mode and population inputs in SKILL.md.\n
  • Boundary markers: None explicitly defined for the ingested research targets.\n
  • Capability inventory: Use of word_export and write_xlsform for document generation.\n
  • Sanitization: Not explicitly defined.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 23, 2026, 05:36 PM
Security Audit — agent-trust-hub — instrument-designer