motivation-diagnostic-task-redesign

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill consists entirely of markdown instructions and metadata. It does not contain any executable code, shell commands, or binary files.
  • [SAFE]: No network operations (curl, wget, fetch) or remote data requests are present in the skill definition.
  • [PROMPT_INJECTION]: The skill accepts several user-controlled inputs, including task descriptions and student profiles, which are interpolated directly into the model prompt. This represents a surface for indirect prompt injection; however, because the skill has no access to sensitive tools, file systems, or network endpoints, any such injection would be confined to the generated text output.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 07:23 AM
Security Audit — agent-trust-hub — motivation-diagnostic-task-redesign