> For the complete documentation index, see [llms.txt](/llms.txt).
> A full single-fetch corpus is available at [llms-full.txt](/llms-full.txt).
---
title: "Scorers"
description: "List and fetch the scorers a project uses to grade experiment results."
last_verified: 2026-07-27
---


**Endpoints**: `GET /api/v1/projects/:projectId/scorers`, `GET /api/v1/projects/:projectId/scorers/:id`
**Auth**: `X-API-KEY` or `Authorization: Bearer <token>` (required)
**Error envelope**: 4xx/5xx return `{"error": string, "message": string, "code": int, "hint"?: string}`


A **scorer** grades a single [experiment](/docs/api-reference/experiments.md) result — deterministic code, an LLM-as-judge prompt, or a built-in like exact-match. This page covers reading scorer definitions; see [Scorers](/docs/improve/scorers.md) for how to write one.

## Next steps

* [Experiments](/docs/api-reference/experiments.md): attach a scorer to an experiment via `scorer_ids`
* [Scorers concept](/docs/improve/scorers.md): write a custom code or LLM-judge scorer
* [Quality Cases](/docs/api-reference/quality-cases.md): track a case a scorer flagged as failing