> ## Documentation Index
> Fetch the complete documentation index at: https://docs.chronicle-labs.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Chronicle

> Grade agents on what they do. Compare versions on repeatable tasks.

Chronicle helps you **grade and compare AI agents** on the work they actually
do. Agents run tasks in a **world**: a simulated system with tools, data, and a
defined starting state. Worldsmith builds that world from your description,
from a single service to a workflow across connected tools.

## How grading works

1. **Define success.** Write a task and attach **scorers** that check its
   requirements. Code checks concrete outputs or recorded changes; an LLM
   judge assesses qualities such as clarity or completeness.
2. **Run and score.** Run tasks in fresh copies of their saved starting world.
   Scorers return a value from **0 to 1**; set pass thresholds to determine
   whether each required check passed.
3. **Compare and improve.** An **evaluation** runs a suite of tasks. Compare
   agent versions on the same tasks and starting state, then inspect the
   recorded actions behind each score to understand failures.

<Card title="Run your first evaluation" icon="play" href="/platform/quickstart">
  Launch an example world, run your agent against its tasks, and inspect the results.
</Card>

## Four things to know

| Concept        | What it is                                                                                                          |
| -------------- | ------------------------------------------------------------------------------------------------------------------- |
| **World**      | A simulated system: the services, data, and state your agent can interact with.                                     |
| **Task**       | A job for the agent, with **scorers** that check the result. A task suite groups jobs to run together.              |
| **Agent**      | The AI agent you want to test.                                                                                      |
| **Evaluation** | Runs agent versions against tasks in fresh worlds and records the scores and evidence. Each attempt is a **trial**. |

## A grading example

The quickstart's duplicate-invoice task has **two checks**, each with a pass
threshold of `1`:

| Check                        | Required evidence                                                         |
| ---------------------------- | ------------------------------------------------------------------------- |
| Engineering started          | ENG-2 changed to **In Progress** during the trial and remains there.      |
| Finance received the handoff | A new message in `#erp-alerts` contains the exact handoff sentence below. |

> ENG-2 is in progress for order 7001; case 00001001 remains open.

The handoff check requires that exact text. Checking that the case actually
stayed open requires a separate scorer; the starter does not include it.

[Tasks and scorers](/platform/tasks-and-scorers) covers writing checks.
[Run an evaluation](/platform/evaluations) explains how to compare versions
and follow scores to their evidence.

## Ways to use Chronicle

<CardGroup cols={2}>
  <Card title="Dashboard" icon="browser" href="/platform/quickstart">
    Build worlds, write tasks, and run evaluations in the browser.
  </Card>

  <Card title="API and SDKs" icon="code" href="/api-reference/overview">
    Automate workflows with Python, TypeScript, JavaScript, Go, or Rust.
  </Card>
</CardGroup>