gap-analysis-from-student-work

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns, hardcoded credentials, or dangerous command executions were detected. The skill is designed for text-based analysis within the agent's safe environment.\n- [PROMPT_INJECTION]: The skill demonstrates a surface for indirect prompt injection through the ingestion of untrusted student work and assessment criteria, though it lacks the capabilities required for exploitation.\n
  • Ingestion points: The variables student_work_description, assessment_criteria, and learning_objective in SKILL.md ingest untrusted text.\n
  • Boundary markers: The prompt uses bold headers to label data but lacks structural delimiters (e.g., XML tags or triple quotes) or specific instructions to disregard instructions within the interpolated text.\n
  • Capability inventory: The skill identifies no tools, shell commands, file system access, or network capabilities.\n
  • Sanitization: No input sanitization or filtering is performed on the user-provided descriptions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 10:53 PM