Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for network engineers

QoS Policy Verification Agent

A tested QoS policy where every class gets the treatment it was designed to get

QoS Policy Verification Agent: what goes in, what the agent does and what you get

What it does

Quality of service policies are set but never verified from end to end. This agent checks the marking, queues and counters along a path. It runs test traffic of different classes, such as voice, video and bulk data, and compares the results with the policy. If drops appear in the wrong class, it traces which device is responsible and proposes a fix. After the engineer approves the change, it runs the same tests again. Edge case: voice is marked correctly at the access switch but remarked to default at the WAN edge, so the agent identifies that device as the fault.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueApprovedYes, continueNoNo 1 STARTS WHEN QoS check is due 2 USES A TOOL Read marking, queue and counter settings along thepath 3 USES A TOOL Send test traffic for each class 4 DOES Compare markings, drops and delay with the policy 5 CHECKS THE RESULT Do all classes meet the policy targets? If not: trace the path to find the device where themarking or queue is wrong. Back to step 2. 6 DOES Propose a fix for the faulty device 7 YOU APPROVE Engineer approves the change 8 USES A TOOL Apply the fix in a window 9 USES A TOOL Rerun the test traffic 10 CHECKS THE RESULT Do all classes now meet the targets? If not: review the next device on the path and proposeanother fix. Back to step 8. 11 RESULT QoS verification report
Read the steps as a list
  1. QoS check is due
  2. Read marking, queue and counter settings along the path
  3. Send test traffic for each class
  4. Compare markings, drops and delay with the policy
  5. Do all classes meet the policy targets?If not: trace the path to find the device where the marking or queue is wrong. Back to step 2.
  6. Propose a fix for the faulty device
  7. Engineer approves the changeThe agent waits here for your OK.
  8. Apply the fix in a window
  9. Rerun the test traffic
  10. Do all classes now meet the targets?If not: review the next device on the path and propose another fix. Back to step 8.
  11. QoS verification report

How it decides

A class passes when markings are preserved along the path and drops and delay stay within the policy targets.

  • Fail a class when markings change along the path
  • Fail a class when delay or drops exceed the target
  • Find the first device where behavior differs
  • Test every class after any fix

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Classes and targets
  • Test paths
  • Test windows
  • Devices included

What keeps you in control

It always asks you first

  • Engineer approves each change

Hard limits

  • Never changes a device without approval
  • Never runs test traffic that disrupts users

It stops when

  • Done: every class passes
  • Stop: the test cannot cover the full path

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensVoice test calls showed 3% loss and 80 ms delay against targets of 1% and 40 ms. The agent traced the markings and found the WAN edge router remarked voice to best effort. The engineer approved a policy fix. The second test showed 0.2% loss and 28 ms delay, and video also passed.

More agents for network engineers