chinese-stem-terminology
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves ingesting and processing untrusted text provided by the user (STEM documents, technical reports, and guidelines) for translation and terminology standardization.
- Ingestion points: Untrusted content enters the agent's context through user-submitted documents and text inputs as defined in the
SKILL.mdworkflow. - Boundary markers: While the skill provides instructions to protect specific structures like code blocks and formulas, it lacks explicit delimiters or instructions to the model to ignore potential malicious prompts embedded within the user's technical text.
- Capability inventory: The skill is granted the capability to rewrite, translate, and modify large blocks of text content, which could be exploited if an injection attack is successful.
- Sanitization: There are no sanitization or input validation mechanisms described for filtering the data processed by the agent.
Audit Metadata