Execution assurance agent guard and observability for autonomous AI agents.

Real-time evaluation of every agent action before it executes — built for agents operating in production.
Built-in collector that captures every tool call as OpenTelemetry metrics and traces.
Works with Claude Code, Codex CLI, OpenClaw, Hermes, Pi, and opencode. More frameworks coming.
Built by Core0 — execution assurance for production AI agents.

→ View the live Execution Pipeline diagram

[](https://agentskills.io)

At a glance

• What it does: Nio hooks into your agent platform (Claude Code, Codex CLI, OpenClaw, Hermes, Pi, opencode) and evaluates each tool call through a Phase 0–6 pipeline before it runs — allow, deny, or confirm — with an optional OTEL collector and local audit log.
• Config: Policy lives in ~/.nio/config.yaml (or $NIO_HOME/config.yaml). Audit events append to ~/.nio/audit.jsonl by default.

Architecture at a glance

Hook events feed the **Collector** (OTEL + local audit) and the **Guard** (real-time Phases 0–6 + on-demand static/behavioural/LLM engines):

┌─────────────────────────────────────────────────────────────────────┐
│                              Nio                                    │
│                                                                     │
│  ┌──────────────────────────┐    ┌───────────────────────────────┐  │
│  │       Collector          │    │            Guard              │  │
│  │                          │    │                               │  │
│  │  Hook events             │    │  ┌─────────────────────────┐  │  │
│  │  → metrics               │