// File generated from our OpenAPI spec by Stainless. See CONTRIBUTING.md for details.

import { asTextContentResult } from 'replicate-mcp/tools/types';

import { Tool } from '@modelcontextprotocol/sdk/types.js';
import type { Metadata } from '../../';
import Replicate from 'replicate-stainless';

export const metadata: Metadata = {
  resource: 'deployments.predictions',
  operation: 'write',
  tags: [],
  httpMethod: 'post',
  httpPath: '/deployments/{deployment_owner}/{deployment_name}/predictions',
  operationId: 'deployments.predictions.create',
};

export const tool: Tool = {
  name: 'create_deployments_predictions',
  description:
    'Create a prediction for the deployment and inputs you provide.\n\nExample cURL request:\n\n```console\ncurl -s -X POST -H \'Prefer: wait\' \\\n  -d \'{"input": {"prompt": "A photo of a bear riding a bicycle over the moon"}}\' \\\n  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \\\n  -H \'Content-Type: application/json\' \\\n  https://api.replicate.com/v1/deployments/acme/my-app-image-generator/predictions\n```\n\nThe request will wait up to 60 seconds for the model to run. If this time is exceeded the prediction will be returned in a `"starting"` state and need to be retrieved using the `predictions.get` endpiont.\n\nFor a complete overview of the `deployments.predictions.create` API check out our documentation on [creating a prediction](https://replicate.com/docs/topics/predictions/create-a-prediction) which covers a variety of use cases.\n',
  inputSchema: {
    type: 'object',
    properties: {
      deployment_owner: {
        type: 'string',
      },
      deployment_name: {
        type: 'string',
      },
      input: {
        type: 'object',
        description:
          "The model's input as a JSON object. The input schema depends on what model you are running. To see the available inputs, click the \"API\" tab on the model you are running or [get the model version](#models.versions.get) and look at its `openapi_schema` property. For example, [stability-ai/sdxl](https://replicate.com/stability-ai/sdxl) takes `prompt` as an input.\n\nFiles should be passed as HTTP URLs or data URLs.\n\nUse an HTTP URL when:\n\n- you have a large file > 256kb\n- you want to be able to use the file multiple times\n- you want your prediction metadata to be associable with your input files\n\nUse a data URL when:\n\n- you have a small file <= 256kb\n- you don't want to upload and host the file somewhere\n- you don't need to use the file again (Replicate will not store it)\n",
      },
      stream: {
        type: 'boolean',
        description:
          '**This field is deprecated.**\n\nRequest a URL to receive streaming output using [server-sent events (SSE)](https://developer.mozilla.org/en-US/docs/Web/API/Server-sent_events).\n\nThis field is no longer needed as the returned prediction will always have a `stream` entry in its `url` property if the model supports streaming.\n',
      },
      webhook: {
        type: 'string',
        description:
          'An HTTPS URL for receiving a webhook when the prediction has new output. The webhook will be a POST request where the request body is the same as the response body of the [get prediction](#predictions.get) operation. If there are network problems, we will retry the webhook a few times, so make sure it can be safely called more than once. Replicate will not follow redirects when sending webhook requests to your service, so be sure to specify a URL that will resolve without redirecting.\n',
      },
      webhook_events_filter: {
        type: 'array',
        description:
          'By default, we will send requests to your webhook URL whenever there are new outputs or the prediction has finished. You can change which events trigger webhook requests by specifying `webhook_events_filter` in the prediction request:\n\n- `start`: immediately on prediction start\n- `output`: each time a prediction generates an output (note that predictions can generate multiple outputs)\n- `logs`: each time log output is generated by a prediction\n- `completed`: when the prediction reaches a terminal state (succeeded/canceled/failed)\n\nFor example, if you only wanted requests to be sent at the start and end of the prediction, you would provide:\n\n```json\n{\n  "input": {\n    "text": "Alice"\n  },\n  "webhook": "https://example.com/my-webhook",\n  "webhook_events_filter": ["start", "completed"]\n}\n```\n\nRequests for event types `output` and `logs` will be sent at most once every 500ms. If you request `start` and `completed` webhooks, then they\'ll always be sent regardless of throttling.\n',
        items: {
          type: 'string',
          enum: ['start', 'output', 'logs', 'completed'],
        },
      },
      Prefer: {
        type: 'string',
      },
    },
  },
};

export const handler = async (replicate: Replicate, args: Record<string, unknown> | undefined) => {
  const body = args as any;
  return asTextContentResult(await replicate.deployments.predictions.create(body));
};

export default { metadata, tool, handler };