skills/bmsuisse/skills/ducklake/Gen Agent Trust Hub

ducklake

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references official documentation and source code hosted on trusted domains, including duckdb.org, github.com/duckdb, and ducklake.select. These are standard technical resources for the tool described.
  • [COMMAND_EXECUTION]: The skill includes SQL commands for managing databases (ATTACH, CREATE TABLE, ALTER TABLE) and a single standard CLI command (duckdb ducklake:...). These are the primary functions of the skill and do not include any suspicious or unauthorized execution patterns.
  • [CREDENTIALS_UNSAFE]: While the skill provides templates for Azure authentication (CREATE SECRET), it correctly uses placeholders and guides the user toward secure practices like using a credential_chain (managed identities). No hardcoded secrets or sensitive files are accessed.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes how to ingest data from external sources (CSV files, Azure blobs). While this represents a data ingestion surface, the instructions are limited to technical database operations with no high-privilege capabilities or evidence of malicious instructions in the processed data templates.
  • [SAFE]: The skill follows security best practices by recommending relative paths for data storage and using official credential providers for cloud access. No obfuscation, persistence, or privilege escalation patterns were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 08:46 AM