IELTS
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFE
Full Analysis
- [DATA_STORAGE]: The skill organizes user study progress, essay samples, and transcripts within a specific local directory (
~/Clawic/data/ielts/). This is a legitimate use of local storage for persistence across study sessions. - [EXTERNAL_DOWNLOADS]: The skill references official and well-known IELTS resources, including the British Council, IDP, and Cambridge University Press. These are established organizations and are treated as safe sources for educational materials.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-generated content such as essays and speaking transcripts for scoring and feedback. While this represents an ingestion surface for untrusted data, the skill lacks dangerous capabilities (such as network exfiltration or arbitrary command execution) that would make this surface exploitable. The instructions include an 'Agent Learning Log' to refine feedback based on user corrections, which is a standard functional behavior for educational agents.
- [SAFE_PRACTICE]: The instructions emphasize adherence to official IELTS band descriptors and exam criteria, ensuring that the agent's behavior remains within the intended educational scope.
Audit Metadata