spark-engineer

Pass

Audited by Gen Agent Trust Hub on Aug 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a set of instructional templates and reference guides for Apache Spark engineering. It contains standard code snippets for PySpark and Scala that follow industry best practices for data processing, partitioning, and performance tuning.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes patterns for ingesting data from external sources, which is the primary function of a Spark-based tool.
  • Ingestion points: SKILL.md (S3 data loading at lines 43, 56) and references/streaming-patterns.md (Kafka and S3 streaming sources at lines 10, 43).
  • Boundary markers: Not present, as the content is instructional code rather than a live prompt system.
  • Capability inventory: The suggested code includes file system writes (S3/Delta Lake) and network connectivity via Spark connectors (Kafka/JDBC).
  • Sanitization: Not applicable to the provided templates.
  • [CREDENTIALS_UNSAFE]: The skill contains example code with placeholders for credentials (e.g., user, password in JDBC examples in references/streaming-patterns.md). These are standard illustrative placeholders and do not represent hardcoded secrets.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 2, 2026, 03:58 PM
Security Audit — agent-trust-hub — spark-engineer