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 Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows established educational frameworks (Brookhart, Sadler, Andrade) and provides clear instructions for the AI to generate text-based rubrics. No malicious patterns such as prompt injection, obfuscation, or data exfiltration were detected.
- [INDIRECT_PROMPT_INJECTION]: The skill contains multiple ingestion points for untrusted user data (e.g.,
{{learning_objective}},{{task_description}}). While it lacks explicit boundary markers to prevent instructions embedded in that data from influencing the agent, the skill has no dangerous capabilities (no file system writes, no network operations, and no command execution). Therefore, the surface area for indirect injection carries no significant risk in this context. - Ingestion points: SKILL.md (Prompt section) interpolates user-provided learning objectives and task descriptions.
- Boundary markers: Absent; the prompt directly interpolates values without delimiters.
- Capability inventory: No tool usage, code execution, or network calls are requested or performed by the skill.
- Sanitization: None detected.
Audit Metadata