disciplinary-writing-scaffold

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a well-structured educational prompt that uses academic research to guide the AI's behavior. No signs of obfuscation, exfiltration, or unauthorized command execution were found.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it ingests multiple untrusted user inputs (e.g., writing tasks and student profiles) without explicit boundary markers. However, the risk is minimal as the skill does not have access to sensitive tools, files, or network operations.
  • Ingestion points: SKILL.md (writing_task, discipline, student_level, genre, student_profiles, assessment_criteria, model_text)
  • Boundary markers: Absent (template variables are interpolated directly into the instructions)
  • Capability inventory: None (the skill uses no external tools, network calls, or file system access)
  • Sanitization: Absent
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 03:33 PM
Security Audit — agent-trust-hub — disciplinary-writing-scaffold