celltypist-cell-annotation
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted biological data files (AnnData, 10x Genomics MTX), which constitutes a potential attack surface for indirect prompt injection.
- Ingestion points:
sc.read_h5adandsc.read_10x_mtxinSKILL.md. - Boundary markers: No specific boundary markers or instructions to ignore embedded content are provided for the processed scientific data.
- Capability inventory: File system access (reading and writing .h5ad files) and package installation via
pip. - Sanitization: No specific sanitization for natural language instructions is implemented for biological count data.
- [EXTERNAL_DOWNLOADS]: Downloads pre-trained logistic regression models from the CellTypist repository for use in cell type prediction.
Audit Metadata