mdf-teaching-script-generator

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a template and instruction set for generating teaching scripts. Comprehensive scanning for obfuscation, privilege escalation, and persistence mechanisms yielded no positive results.- [PROMPT_INJECTION]: The instructions contain strict operational constraints for the AI (e.g., "prohibit delivery" on errors) and formatting requirements, but these are task-specific and do not target the underlying safety guardrails of the model.- [DATA_EXFILTRATION]: There are no network requests to untrusted domains or attempts to access local sensitive files. All referenced external URLs are documentation links for the MarkdownFlow specification.- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied course fragments. While this is an ingestion surface for untrusted data, the skill includes explicit boundary markers (===...===) and structural delimiters (---) that help differentiate between control instructions and data content, mitigating the risk of the model following instructions embedded within the source materials.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 09:20 AM