Skip to content

docs: standardize landing page CLI/Web wording #4

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
sebastianmaniak merged 1 commit into from
Mar 22, 2026
Merged

docs: standardize landing page CLI/Web wording #4

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
210 changes: 91 additions & 119 deletions layouts/index.html
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,141 +1,113 @@
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>{{ .Site.Title }}</title>
{{ partial "head.html" . }}
</head>
<body>
<div class="bg-grid"></div>

<header class="hero">
<div class="hero-content">
<div class="hero-badge">
<span class="badge-dot"></span>
Open source • Python SDK • OpenTelemetry native
</div>
<h1>Score your AI agent behavior from traces.</h1>
<p class="hero-subtitle">
AgentEvals is the open-source Python framework for scoring AI agent performance and behavior
from OpenTelemetry traces. Test prompts, tools, memory, and workflows without re-running your agents.
</p>
<div class="hero-cta">
<a href="/docs/quick-start/" class="btn btn-primary">Quick Start</a>
<a href="https://github.com/agentevals-dev/agentevals" class="btn btn-secondary" target="_blank" rel="noopener">GitHub</a>
</div>
<div class="hero-meta">
<span>CLI</span>
<span>Custom Evaluators</span>
<span>Web UI</span>
<span>CI/CD</span>
{{ define "main" }}
<section class="hero">
<div class="container hero-grid">
<div>
<p class="eyebrow">OpenTelemetry-native agent evaluation</p>
<h1>Score AI agents from traces — no reruns required</h1>
<p class="lead">
agentevals turns OpenTelemetry traces into repeatable, rubric-based scores for tool use,
handoffs, planning, and other agent behaviors.
</p>
<div class="hero-actions">
<a class="btn btn-primary" href="/docs/quick-start/">Start with the CLI</a>
<a class="btn btn-secondary" href="/docs/ui-walkthrough/">Open Web</a>
</div>
</div>
<aside class="hero-card">
<div class="terminal">
<div class="terminal-bar">
<span></span><span></span><span></span>
</div>
<pre><code>$ uv tool install agentevals
$ agentevals score traces.jsonl \
--config agentevals.yaml

✔ 184 traces scored
✔ 91% tool-call success
✔ Mean rubric score: 4.4 / 5.0</code></pre>
</div>
</aside>
</div>
</header>
</section>

<main>
<section class="features section">
<section class="section">
<div class="container">
<div class="section-header">
<span class="section-label">Why AgentEvals</span>
<h2>Evaluation that matches how agents actually run.</h2>
<p>Traditional evals re-run entire workflows. AgentEvals scores the traces you already collect, so you can measure behavior in realistic conditions.</p>
<p class="eyebrow">Why agentevals</p>
<h2>Evaluate behavior from the telemetry you already collect</h2>
<p>
Score agents against consistent rubrics using OpenTelemetry traces rather than replaying runs.
Keep evaluations close to your production workflows and compare changes over time.
</p>
</div>

<div class="feature-grid">
<article class="feature-card">
<div class="feature-icon">◉</div>
<h3>Trace-native evaluation</h3>
<p>Built on OpenTelemetry traces so you can evaluate real production-like runs without replaying agent execution.</p>
<h3>No reruns</h3>
<p>Use recorded traces to evaluate real executions after the fact.</p>
</article>
<article class="feature-card">
<div class="feature-icon">◈</div>
<h3>Flexible scoring</h3>
<p>Combine built-in evaluators with custom Python logic to measure correctness, tool usage, memory behavior, and more.</p>
<h3>Behavior-first scoring</h3>
<p>Measure task completion, tool use quality, handoffs, latency, and more.</p>
</article>
<article class="feature-card">
<div class="feature-icon">◎</div>
<h3>Works in your workflow</h3>
<p>Run locally with the CLI, automate in CI/CD, or explore results visually in the web UI.</p>
<h3>Built on OpenTelemetry</h3>
<p>Plug into existing observability pipelines instead of inventing a parallel eval stack.</p>
</article>
</div>
</section>
</div>
</section>

<section class="workflow section">
<section class="section alt">
<div class="container">
<div class="section-header">
<span class="section-label">How it works</span>
<h2>From traces to scores in three steps.</h2>
<p class="eyebrow">How it works</p>
<h2>Two ways to evaluate</h2>
</div>

<div class="workflow-steps">
<div class="workflow-step">
<span class="step-number">01</span>
<h3>Collect traces</h3>
<p>Instrument your agent with OpenTelemetry and emit traces for prompts, tool calls, memory operations, and outputs.</p>
</div>
<div class="workflow-step">
<span class="step-number">02</span>
<h3>Define evaluators</h3>
<p>Choose built-in evaluators or create your own to score the behaviors that matter for your agent.</p>
</div>
<div class="workflow-step">
<span class="step-number">03</span>
<h3>Run evaluations</h3>
<p>Score trace datasets through the CLI or web UI and compare results across prompts, models, or tool strategies.</p>
</div>
<div class="steps-grid two-up">
<article class="step-card">
<span class="step-number">1</span>
<h3>CLI workflow</h3>
<p>
Run evaluations locally or in CI with config files and reproducible commands.
</p>
<a href="/docs/quick-start/">Open the CLI guide →</a>
</article>
<article class="step-card">
<span class="step-number">2</span>
<h3>Web workflow</h3>
<p>
Explore traces, inspect scores, and review rubric results in the browser.
</p>
<a href="/docs/ui-walkthrough/">Open the Web guide →</a>
</article>
</div>
</section>
</div>
</section>

<section class="docs-preview section">
<section class="section">
<div class="container">
<div class="section-header">
<span class="section-label">Docs</span>
<h2>Start with the path that fits your workflow.</h2>
<p class="eyebrow">Docs</p>
<h2>Start where you are</h2>
</div>

<div class="docs-grid">
{{ range where .Site.RegularPages "Section" "docs" }}
<a class="doc-card" href="{{ .RelPermalink }}">
<div>
<h3>{{ .Title }}</h3>
<p>{{ .Description }}</p>
</div>
<span class="doc-arrow">→</span>
</a>
{{ end }}
<a class="docs-card" href="/docs/quick-start/">
<h3>Quick start</h3>
<p>Install agentevals, run your first scoring pass, and inspect the output.</p>
</a>
<a class="docs-card" href="/docs/integrations/">
<h3>Integrations</h3>
<p>Connect agentevals with your existing tracing and observability stack.</p>
</a>
<a class="docs-card" href="/docs/custom-evaluators/">
<h3>Custom evaluators</h3>
<p>Define your own scoring logic and tailor rubrics to your agents.</p>
</a>
<a class="docs-card" href="/docs/ui-walkthrough/">
<h3>Web walkthrough</h3>
<p>See how to inspect traces and scores with the browser-based interface.</p>
</a>
</div>
</section>

<section class="usage section">
<div class="section-header">
<span class="section-label">Usage</span>
<h2>Two ways to evaluate.</h2>
<p>Use the CLI for fast, scriptable scoring or the Web UI for visual exploration of evaluation results.</p>
</div>

<div class="usage-grid">
<article class="usage-card">
<h3>CLI</h3>
<p>Run evaluations locally or in CI with straightforward commands and structured outputs.</p>
<pre><code>agentevals eval run config.yaml</code></pre>
</article>
<article class="usage-card">
<h3>Web UI</h3>
<p>Inspect trace datasets, compare runs, and review evaluator outputs in a visual interface.</p>
<pre><code>agentevals ui</code></pre>
</article>
</div>
</section>

<section class="cta section">
<div class="cta-card">
<span class="section-label">Get started</span>
<h2>Bring evaluation into your agent development loop.</h2>
<p>Install AgentEvals, connect your traces, and start measuring how your agent behaves in the real world.</p>
<div class="hero-cta">
<a href="/docs/quick-start/" class="btn btn-primary">Read the docs</a>
<a href="https://github.com/agentevals-dev/agentevals" class="btn btn-secondary" target="_blank" rel="noopener">View on GitHub</a>
</div>
</div>
</section>
</main>
</body>
</html>
</div>
</section>
{{ end }}
Loading