ml4t-feature-store
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides patterns for processing and storing external datasets, which creates a potential surface for indirect injection if malicious data is ingested.
- Ingestion points: The
save_featuresandload_featuresfunctions inSKILL.mdingestpolars.DataFrameobjects which are typically sourced from external files or APIs. - Boundary markers: The skill does not explicitly define LLM boundary markers or delimiters to isolate processed data from agent instructions.
- Capability inventory: The provided code snippets perform file system operations including directory creation (
pathlib.Path.mkdir) and data persistence (polars.DataFrame.write_parquet). - Sanitization: The skill includes a
validate_schemafunction that enforces required columns and checks for null values, providing a layer of data integrity validation.
Audit Metadata