snowflake-expert

Pass

Audited by Gen Agent Trust Hub on Jun 28, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill provides patterns for processing data from external cloud storage and internal streams, establishing a surface for indirect prompt injection.
  • Ingestion points: The SKILL.md file contains SQL templates for CREATE EXTERNAL TABLE and COPY INTO, which enable the agent to read data from external S3 and Azure Blob Storage locations.
  • Boundary markers: Absent. The provided SQL snippets do not include instructions or delimiters to help the agent distinguish between trusted instructions and untrusted data from these external sources.
  • Capability inventory: The skill is configured with access to powerful tools including Bash, Write, and Edit in its frontmatter.
  • Sanitization: Absent. The examples focus on data platform functionality and do not demonstrate validation or escaping techniques for ingested content.
  • [SAFE]: The skill uses safe placeholders such as 'xxx' and 'yyy' for sensitive credentials in its Snowflake configuration examples.
  • [SAFE]: The dependency snowflake-connector-python is an official and well-known library for Snowflake integration.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 28, 2026, 05:51 PM
Security Audit — agent-trust-hub — snowflake-expert