Zentinelle

Runtime governance for AI coding agents.

Zentinelle sits between your AI agents and their LLM providers, enforcing policies, logging activity, and giving you a real-time compliance dashboard.

Alpha

v0.0.1 — Working integrations with Claude Code and Codex. Looking for contributors.

How it works

Your Agent  ──►  Zentinelle  ──►  LLM Provider
               (policy check)    (Anthropic, OpenAI, Google)
               (audit log)
               (monitoring)

Two integration modes:

Mode	How	Best for
Hooks	PreToolUse/PostToolUse intercepts	Claude Code, Gemini CLI
Proxy	Transparent HTTP proxy	Any agent (Codex, custom)

Quick links

• Quick Start -- get running in 5 minutes
• Architecture -- how the system is built
• API Reference -- REST + GraphQL endpoints
• Policy Reference -- all policy types and configs
• SDK Guide -- integrate your agent
• Knowledge Graph -- codebase architecture map
• Contributing