naming-tagging-standards

Pass

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its flexible configuration resolution logic. It is explicitly instructed to scan the context/ directory for files in any format (YAML, CSV, Markdown, JSON, TXT, or free-form text) and "interpret intent" to derive tagging standards and smart defaults.
  • Ingestion points: The agent scans the context/ directory for files matching naming patterns like *tag*, *tagging*, or *governance* (defined in SKILL.md).
  • Boundary markers: No delimiters or instructions to ignore embedded commands are specified when parsing these external files.
  • Capability inventory: The skill generates and executes SQL commands (ALTER CATALOG, ALTER SCHEMA, ALTER TABLE ... SET TAGS) and modifies Databricks Asset Bundle configurations based on the parsed data.
  • Sanitization: There is no evidence of sanitization or strict schema validation for free-form or natural language inputs extracted from the user-provided files.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 8, 2026, 02:33 AM