Skip to main content
Manage evaluation results from the PraisonAI CLI.

​
Quick Start

1

Simple Usage

npx praisonai eval results summary results.json
2

With Configuration

npx praisonai eval results export results.json --format csv

​
Eval Results CLI

Manage and analyze Agent evaluation results from the command line.

​
View Results

# List all results
npx praisonai eval results list

# Filter by status
npx praisonai eval results list --passed
npx praisonai eval results list --failed

# Filter by category
npx praisonai eval results list --category math

​
Aggregate Stats

# Show summary
npx praisonai eval results summary

# Show by category
npx praisonai eval results summary --by-category

# Show trend
npx praisonai eval results trend --window 1h

​
Format Output

# Table format
npx praisonai eval results --format table

# JSON format
npx praisonai eval results --format json

# Markdown format
npx praisonai eval results --format markdown

​
Export Results

# Export to JSON
npx praisonai eval results export --output results.json

# Export to CSV
npx praisonai eval results export --format csv --output results.csv

​
Import Results

# Import from file
npx praisonai eval results import results.json

# Merge with existing
npx praisonai eval results import new-results.json --merge

​
Programmatic (TypeScript)

import { EvalResults, createEvalResults } from 'praisonai';

const results = createEvalResults();
results.add({ name: 'test', passed: true, score: 0.9, duration: 100 });
console.log(results.formatSummary());

Eval Results

SDK result collection

Evaluation CLI

Run evaluations