content-translator
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a secure tool-calling pattern by requiring a discovery step (
sandbase_describe_tool) to confirm schemas before tool execution (sandbase_call_tool). All tools used (strale_translate) are vendor-specific resources provided by sandbaseai. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external text content for translation, which constitutes an ingestion surface for indirect prompt injection. However, the risk is minimal as the skill's capabilities are scoped to content transformation.
- Ingestion points: Processes user-provided text content for translation tasks (SKILL.md).
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are defined for the input text.
- Capability inventory: The skill is limited to calling the
strale_translatetool via the SandBase gateway; it does not possess file-writing, network exfiltration, or shell execution capabilities. - Sanitization: No explicit sanitization or filtering of the input text is mentioned in the instructions.
Audit Metadata