GouvernAI plugin is permissive with your AI agent where risk is low. Conservative where it matters. Invisible everywhere else.

GouvernAI is a Claude Code plugin with two enforcement layers. ~60% of typical actions (reads, drafts, navigation) flow through with zero friction. Guardrails only activate when risk is real.

Skill layer: classifies actions across 4 risk tiers — T1 excluded, T2 notifies, T3 pauses for approval, T4 requires full stop with risk assessment. Escalation rules bump tiers for bulk operations, unfamiliar targets, and scope expansion.

Hook layer: PreToolUse hooks run on every Bash, Write, and Edit call. Hard-blocks obfuscated commands, credential transmission, catastrophic operations, and self-modification. No override.

Includes /guardrails slash command with persistent modes (strict, relaxed, audit-only), audit-only mode for autonomous agents, and append-only audit logging. 85 unit tests. Zero dependencies. MIT license.
Website: https://gouvernai.ai

Features

  • Dual enforcement (skill + hooks)
  • 4-tier risk classification
  • Audit logging
  • Credential exfiltration blocking
  • Obfuscated command detection
  • AI agent
  • Claude Code
  • automode

Project Activity

See All Activity >

Follow GouvernAI — Claude Code Plugin

GouvernAI — Claude Code Plugin Web Site

Other Useful Business Software
Gemini 3 and 200+ AI Models on One Platform Icon
Gemini 3 and 200+ AI Models on One Platform

Access Google's best plus Claude, Llama, and Gemma. Fine-tune and deploy from one console.

Build generative AI apps with Vertex AI. Switch between models without switching platforms.
Start Free
Rate This Project
Login To Rate This Project

User Reviews

Be the first to post a review of GouvernAI — Claude Code Plugin!

Additional Project Details

Registered

2026-03-19
Report inappropriate content