vitess

Pass

Audited by Gen Agent Trust Hub on Oct 24, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: Documentation and Architecture: The skill provides educational content and architectural overviews for the Vitess database system, aligning with its stated purpose of assisting with database management.
  • [SAFE]: External Documentation: The skill references external markdown files for supplemental guides. These are hosted on the official PlanetScale GitHub organization and contain legitimate technical documentation.
  • [SAFE]: Administrative Command Documentation: The skill documents the use of standard database administrative tools such as vtctldclient and SQL session variables for operational tasks like monitoring migrations and query troubleshooting.
  • [INDIRECT_PROMPT_INJECTION]: Data Ingestion Surface: The skill facilitates interaction with database systems where the agent may process external data from query results or configuration files.
  • Ingestion points: The agent context may ingest untrusted data from database query results (SELECT), VSchema JSON definitions, and migration status reports via vtctldclient commands.
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the documentation.
  • Capability inventory: The skill involves executing administrative shell commands (vtctldclient) and SQL queries via the VTGate proxy.
  • Sanitization: No specific sanitization or validation logic is provided within the documentation itself.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 24, 2026, 04:12 PM