Skip to content

Laya Guard 0.6 adds session auditing and a web trial

Updated Laya guard

Summary

An audit command replays prior shell commands through guard policies; a web trial shows verdicts, while 0.6.1 fixes force-push recognition.

Date
Pricing
Not stated in the sources we cited.

Laya Guard published v0.6.0 and v0.6.1 on October 23, 2026.

Audit before installing a hook

Version 0.6.0 adds laya-guard audit, which reads supported Claude Code or Codex transcripts and evaluates recorded shell commands as the corresponding hook would. It groups repeated commands and lists reviews and blocks, with time-window, agent, redaction and JSON options.

The release also changes model-only blocks into permission reviews where an adapter can ask the user. Protected-path and block-level pattern rules still apply. This is a policy-behavior change, not proof that every dangerous command is detected.

Trial and force-push fix

Version 0.6.1 adds a browser trial showing verdicts and decision steps for selected agent hooks. Its release notes distinguish the trial from the configurable local dashboard, which must remain on localhost.

The patch also recognizes additional force-push forms, including -f and plus-prefixed refspecs, as reviews. Review adapter behavior and data access before using transcript auditing. See the Laya Guard hub.

More from Laya guard

View all

Questions

What is Laya guard?

According to its repository description, it is a free local guardrail for AI coding agents that judges every tool_call on-device (regex + Laya) as allow, review or block.

Is Laya guard open source?

The GitHub API reports an MIT license for the repository.

Filed on Agent Radar. Claims in this note are limited to the sources above.