arken-prepare-case-draft
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill transmits user-provided context to the official ARKEN API endpoint (usearken.com). This behavior is consistent with the skill's primary purpose and targets the official domain of the publisher.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied context, creating a potential surface for indirect prompt injection. To mitigate this risk, the instructions explicitly require the agent to remove credentials and unrelated personal data before the data is sent to the API.
Audit Metadata