tanstack-db

Pass

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references official library packages from the TanStack organization and provides instructions for fetching related agent skills from the vendor's own repository (oakoss/agent-skills). These sources are well-known and consistent with the skill's purpose.
  • [COMMAND_EXECUTION]: Documentation includes standard shell commands (npm, pnpm, npx) for package installation and agent skill expansion. These are routine development operations.
  • [SAFE]: The skill uses best practices for data handling, such as promoting Zod schema validation for external data sources to prevent malformed data ingestion.
  • [PROMPT_INJECTION]: The skill possesses a data ingestion surface through live queries and API sync handlers; however, it mitigates potential risks by encouraging strict schema validation (e.g., in references/local-and-storage-collections.md and references/setup.md).
  • Ingestion points: Data enters the system via 'queryFn' in queryCollectionOptions and 'shapeOptions' in electricCollectionOptions.
  • Boundary markers: Data is processed through a structured reactive database engine with defined query parameters.
  • Capability inventory: The skill uses 'fetch' for network operations and 'localStorage' for browser persistence.
  • Sanitization: Explicitly recommends the use of Zod-based schemas for runtime validation of incoming data.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 25, 2026, 06:52 AM