evaluator

Automate agent evaluation metric runs and job monitoring in NeMo Platform.

58|16|Updated May 14, 2026
One-click install
npx skills add https://github.com/NVIDIA-NeMo/nemo-platform --skill evaluator-nvidia-nemo
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: evaluator
Source: https://github.com/NVIDIA-NeMo/nemo-platform/tree/main/agents/nemo-agent-local/src/nemo_agent/skills/evaluator
Command: npx skills add https://github.com/NVIDIA-NeMo/nemo-platform --skill evaluator-nvidia-nemo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill addresses the complexity of managing agent performance metrics, allowing users to automate the evaluation lifecycle from dataset creation to asynchronous job monitoring.

Core Features & Use Cases

  • Metric Lifecycle Management: Orchestrates the creation of workspaces, datasets, and metrics through the NeMo Platform SDK.
  • Synchronous and Asynchronous Execution: Supports both immediate inline metric evaluation and long-running background metric jobs.
  • Use Case: A developer can use this Skill to automatically trigger a string-check evaluation on a new agent version, monitor the job status, and retrieve results without manual CLI intervention.

Quick Start

Use the evaluator skill to create a new metric job for the current workspace and monitor its status until completion.

Frequently Asked Questions about evaluator

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

FAQPage Schema
How do I automate agent evaluation metrics in NeMo Platform?

You can automate agent evaluation metrics by using the NeMo Platform SDK to orchestrate workspace creation, dataset preparation, and metric execution without manual CLI intervention. This streamlines the entire metric lifecycle from start to finish.

What is asynchronous metric job monitoring in NeMo?

Asynchronous metric job monitoring allows you to trigger long-running background evaluation jobs and poll their status using the nemo_api evaluation module. This enables you to retrieve results later without blocking your development environment.

Can I run synchronous inline metric evaluations for agent performance?

Yes, synchronous execution supports immediate inline metric evaluation for quick checks. This allows you to perform rapid string-check evaluations on new agent versions and get instant feedback without waiting for background processing.

Do I need the nemo_api evaluation module to manage metric jobs?

Yes, integration with the nemo_api evaluation module is required to handle metric operations and job status polling. This dependency allows the automation process to manage both synchronous and asynchronous evaluation jobs effectively.

What are the limitations of automating agent evaluation with this approach?

The primary limitation is that metric lifecycle management is tightly coupled to the NeMo Platform environment. Users must operate within this specific ecosystem to successfully execute synchronous and asynchronous evaluation jobs.