The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external, potentially untrusted sources like S3 buckets, Kafka topics, and JDBC databases. It lacks instructions for data sanitization or the use of boundary markers to separate data from instructions, which is a common surface for indirect prompt injection attacks.
Ingestion points: Untrusted data enters the context via spark.read.parquet("s3://...") in SKILL.md and through Kafka/File sources in references/streaming-patterns.md.
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are provided for the agent when handling the content of these data sources.
Capability inventory: The skill possesses the ability to write to the file system (S3/Local) and perform network operations through Kafka and JDBC connections.
Sanitization: The provided code examples do not include logic for escaping, validating, or filtering external content before it is processed or written.
[EXTERNAL_DOWNLOADS]: The skill's documentation link https://jeffallan.github.io/claude-skills/skills/data-ml/spark-engineer/ has been flagged as malicious by automated URL scanners. Although the domain is a well-known service (GitHub Pages), the specific content within this path is identified as a risk by external security intelligence.

spark-engineer