aidp-object-storage

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime reads/writes OCI data via oci://... paths (Spark read.load / df.write.save and CREATE EXTERNAL VOLUME/TABLE ... LOCATION/path), so any file contents in the user-specified bucket become LLM-readable text in the agent context (e.g., CSV/JSON/Parquet converted to strings), and those files are outsider-authored relative to the operating user.