goodcase
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates as a standard data-retrieval tool, using
curlto fetch JSON data from a documented public endpoint. - [EXTERNAL_DOWNLOADS]: Fetches case study data and prompt strings from
https://goodcase.ai/api/public. These are legitimate data requests for the skill's stated purpose. - [INDIRECT_PROMPT_INJECTION]: As the skill ingests raw AI prompts from an external source (
promptFullfield), it possesses an attack surface for indirect prompt injection if the remote data contains adversarial instructions. This is an inherent risk of its functionality.
Audit Metadata