omniroute-monitoring

Retrieve OmniRoute system health, circuit breaker states, and per-provider latency metrics.

44.0k|5.9k|Updated Feb 13, 2026
One-click install
npx skills add https://github.com/diegosouzapw/OmniRoute --skill omniroute-monitoring
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: omniroute-monitoring
Source: https://github.com/diegosouzapw/OmniRoute/tree/main/skills/omniroute-monitoring
Command: npx skills add https://github.com/diegosouzapw/OmniRoute --skill omniroute-monitoring

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you quickly detect whether OmniRoute is healthy and which providers are slowing down or failing, while also controlling spend with quota and budget limits.

Core Features & Use Cases

  • System health checks: Verify uptime, memory, active connections, circuit breaker states, rate-limit status, and cache statistics to confirm the gateway is functioning.
  • Provider circuit breaker visibility: Inspect per-provider states (CLOSED, OPEN, HALF_OPEN) and understand when recovery will occur.
  • Per-provider performance & usage metrics: Track p50/p95/p99 latency, success rate, tokens used, and requests for pinpoint debugging and optimization.
  • Quota and spend governance: Check quota balances and enforce budget guardrails to degrade, block, or alert when limits are hit.
  • MCP monitoring and audit visibility: Retrieve MCP server status and a summary of recent MCP tool-call audit activity for operational oversight.

Quick Start

Ask your AI to check OmniRoute health and provider metrics by calling the health and metrics endpoints using your OMNIROUTE_URL and OMNIROUTE_KEY.

Frequently Asked Questions about omniroute-monitoring

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I monitor AI gateway health and circuit breaker states?

AI gateway health monitoring retrieves system uptime, memory, active connections, and circuit breaker states (CLOSED, OPEN, HALF_OPEN) to identify failing providers. It also tracks recovery timing using authenticated access to your gateway URL and API key.

How can I track p50 p95 p99 latency for API providers?

Tracking p50, p95, and p99 latency involves retrieving per-provider performance metrics including success rates, tokens used, and request counts. This provides the structured data needed to pinpoint performance bottlenecks during production debugging.

How do I set up quota usage and budget guard alerts for an AI gateway?

Setting up quota usage and budget guards checks quota balances and enforces budget guardrails. This allows you to degrade, block, or alert automatically when spending limits are hit, ensuring spend governance across providers.

Can I audit MCP tool calls and server status during production troubleshooting?

You can audit MCP tool calls by retrieving MCP server status and a summary of recent tool-call audit activity. This provides operational oversight for MCP monitoring and confirms whether the MCP servers are functioning correctly.

What do I need to check OmniRoute provider metrics and usage?

Checking OmniRoute provider metrics requires authenticated access via OMNIROUTE_URL and OMNIROUTE_KEY. Once configured, you can retrieve structured health, metrics, quota, and budget guard responses for oncall-style oversight.