Skip to content

1. Home

Jump to bottom
Hemant Kohli edited this page Sep 23, 2026 · 4 revisions

Genkit Guard

Try the Hugging Face Space with synthetic inputs.

Genkit Guard checks injection patterns and semantic intent, masks detected personal data before model calls, restores tokens in responses and tool inputs, and enforces tool policies. Detection can miss attacks or PII and reject valid requests; keep authorization in your application.

These pages describe the v0.2.0 codebase being prepared for release. Documentation is maintained in a single docs/wiki folder.

Start with 3.-Getting-Started or 6.-Full-Setup-Guide. See 8.-Shared-Model-Configuration for reusable configuration, 9.-Tool-Controls-and-Logging for tool policies, and 10.-Decision-Storage-and-Model-Fallback for audit persistence and local model fallback.

New in v0.2.0: 11.-Framework-Compatibility and 12.-Security-and-Operational-Errors cover the compatibility matrix, security hardening and sanitized failures.

Release plan

  • v0.1.0: persistent decision logging and explicit local model fallback.
  • v0.2.0: framework compatibility, security hardening and operational error handling.
  • v1.0.0-rc.1 (next): release candidate and public API freeze for exports, configuration, decision schema and error codes. Subsequent candidates focus on fixes and validation; API changes require explicit review before stable v1.
  • SQLite vault and memory compaction are deferred beyond this release candidate scope.

Clone this wiki locally