spark
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate technical documentation, best practices, and code examples for distributed data processing using Apache Spark and PySpark.
- [INDIRECT_PROMPT_INJECTION]: Ingestion points: Data is read from external sources such as S3 and Kafka via
spark.read.parquetandspark.readStreaminSKILL.md. Boundary markers: No explicit delimiters or instructions to ignore embedded content are present in the examples. Capability inventory: Capabilities are focused on distributed data processing and writing results back to cloud storage; no direct OS command execution or local file system writes are present. Sanitization: No data sanitization is shown as the examples focus on analytical transformations. This represents a standard data processing surface and is considered safe.
Audit Metadata