ml4t-build-bars
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements standard financial data engineering patterns for aggregating trade data. Analysis of the code and instructions revealed no signs of malicious intent, unauthorized data access, or obfuscation.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external trade data (e.g., parquet files).
- Ingestion points: The code examples ingest data via
pl.read_parquet("trades.parquet"). - Boundary markers: None present; the skill expects structured trade data with specific columns (timestamp, price, size).
- Capability inventory: Data processing and aggregation using Polars, Numpy, and the
ml4t-engineerlibrary. - Sanitization: No sanitization is applied to the raw trade data, which is standard practice for numerical financial datasets.
- [EXTERNAL_DOWNLOADS]: The skill references the
ml4t-engineerlibrary, which is a library provided by the skill author for financial engineering tasks.
Audit Metadata