---
title: "Get Eval Run Drafts"
method: GET
path: "/api/platform/v1/eval-runs/{id}/drafts"
tags: ["observability"]
---

# Get Eval Run Drafts

`GET /api/platform/v1/eval-runs/{id}/drafts`

## Request

- Base URL: `https://docs.amdahl.ai`
- URL: `https://docs.amdahl.ai/api/platform/v1/eval-runs/{id}/drafts`
- Auth: one of:
  - API key in header `X-API-Key`
  - HTTP bearer

Fetch both graded artifacts of one eval run side by side: the prompt and message you submitted, the improved versions the eval wrote, each with its score, checks fraction and per-rubric-line reasoning. Read when comparing your draft against the rewrite line by line — this is the only read that carries the texts themselves. Cited quotes stay on the evidence read.

## Path parameters

- `id` string, required

## Response `200`

OK

- object
  - `data` object, required
    - `run_id` string, required
    - `eval_slug` string, required
    - `eval_version` string, nullable
    - `status` 'queued' | 'running' | 'complete' | 'failed' | 'canceled', required: The run status, at the root as on every eval read. The terminal success value is `complete` — never `completed`.
    - `submitted` object: The prompt and message you sent, each with its score and per-line verdicts.
    - `improved` object: The rewritten versions. `prompt` / `message` are null with an `absent_reason` on a run that correctly wrote no rewrite — never an empty string, never a zero score.

## Other responses

- `400`: Invalid input - failed schema validation
- `401`: Authentication required
- `403`: Forbidden - missing scope or role
- `404`: Not found
- `500`: Internal server error

---

[API](https://skmtc.dev/amdahl/apis/amdahl-platform-api.md) · [All operations](https://skmtc.dev/amdahl/apis/amdahl-platform-api/llms.txt) · [OpenAPI document](https://skmtc.dev/amdahl/apis/amdahl-platform-api/revisions/d4eee33e2472?raw)