skills/garethmanning/education-agent-skills/criterion-referenced-rubric-generator/Gen Agent Trust Hub
criterion-referenced-rubric-generator
Pass
Audited by Gen Agent Trust Hub on May 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of a single Markdown file providing instructional text and a prompt template. It contains no executable scripts, binaries, or code dependencies.
- [PROMPT_INJECTION]: The skill ingests user data via placeholders including 'learning_objective', 'task_description', and 'existing_criteria'. While these represent an indirect prompt injection surface, the risk is mitigated by the total absence of exploitable capabilities such as network access or shell execution. The prompt itself contains clear instructions focused on educational assessment and does not attempt to bypass agent safety guidelines.
- [DATA_EXFILTRATION]: There are no tools or commands for network communication or sensitive file access. No credentials or secret management patterns are present.
- [REMOTE_CODE_EXECUTION]: The skill does not download external code or scripts and does not use package managers like npm or pip.
- [COMMAND_EXECUTION]: No shell commands or system-level interactions are defined in the skill instructions or metadata.
Audit Metadata