motherduck-duckdb-sql

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as a documentation reference. While it includes examples of SQL commands that interact with the filesystem (e.g., COPY TO) and remote storage (e.g., read_parquet from S3), these are standard features of the DuckDB/MotherDuck platform and are presented within a controlled instructional context. No active scripts or tools were found that would execute these commands without user intervention.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and repair user-provided SQL statements. While this represents a potential injection surface, the skill's primary function is informational syntax lookup. The instructions do not grant the agent high-privilege autonomous capabilities that would be susceptible to malicious payload execution via this surface.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 09:19 PM