excel-generator
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data (retrieved via SandBase tools) and user-provided files to generate and verify workbooks, creating a vector for indirect prompt injection.
- Ingestion points: Data is ingested from external sources through the SandBase API (e.g., via
sandbase_run) and from existing workspace files as detailed inSKILL.md. - Boundary markers: The instructions do not define specific delimiters or warnings to ignore instructions that might be embedded within the retrieved data.
- Capability inventory: The skill possesses capabilities for writing files (
.xlsx) and conducting network operations via designated vendor tools. - Sanitization: There are no explicit instructions to sanitize or validate the content of the retrieved data for malicious instructions before it is processed or written into the workbook.
- [EXTERNAL_DOWNLOADS]: The skill utilizes vendor-specific tools (
sandbase_discover,sandbase_inspect,sandbase_run) provided by the author, sandbaseai, to retrieve structured external data. These operations include transparency regarding price and call counts, requiring explicit user approval before execution.
Audit Metadata