---
title: "Screen a prompt"
method: POST
path: "/v1/moderation/prompt"
tags: ["Moderation"]
---

# Screen a prompt

`POST /v1/moderation/prompt`

Evaluate a text prompt against content policies before generation. This endpoint is experimental and may change.

## Request body

- ScreenPromptRequest
  - `prompt` string, required — The text prompt to evaluate against content policies.
  - `external_id` string — An optional identifier to associate this request with.

## Response `200`

Prompt screening result

- ScreenPromptResponse
  - `id` string, required — Unique identifier for the moderation result.
  - `object` string, required — Object type.
  - `prompt` string, required — The prompt that was screened.
  - `external_id` string — The external identifier provided in the request.
  - `decision` 'allow' | 'deny' | 'flag', required — The moderation decision.
  - `usage` UsageEntity, required
    - `units` number, required — Number of units consumed by this call.

## Other responses

- `400` — Bad Request - Invalid input parameters
- `401` — Unauthorized - Invalid or missing API key
- `404` — Not Found - Resource does not exist

---

[API](https://skmtc.net/creem/apis/creem-api.md) · [All operations](https://skmtc.net/creem/apis/creem-api/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/creem/creem-api/revisions/aa593160d96c/schema)