> ## Documentation Index > Fetch the complete documentation index at: https://docs.taidalos.com/llms.txt > Use this file to discover all available pages before exploring further. # Create Configuration ## OpenAPI ````yaml POST /v1/media-searches/configuration/create-configuration openapi: 3.0.0 info: title: Taidalos API Documentation termsOfService: https://www.taidalos.com/gtc/ contact: name: Customer Support url: https://www.taidalos.com email: cs@taidalos.com version: '1.0' servers: - url: https://api.taidalos.com security: [] tags: - description: Operations about subjects name: Subjects - description: Operations about media search configuration name: Media Search Configuration - description: Operations about media search name: Media Search externalDocs: description: Get started with guides url: https://taidalos.com/tags/guide/ paths: /v1/media-searches/configuration/create-configuration: post: tags: - Media Search Configuration summary: Create Media Search Configuration parameters: - description: name: Authorization in: header required: true schema: type: string requestBody: content: application/json: schema: $ref: '#/components/schemas/main.createMediaSearchConfigRequest' description: request body required: true responses: '201': description: Created content: application/json: schema: $ref: '#/components/schemas/main.createSearchConfigResponse' '400': description: Bad Request content: application/json: schema: $ref: '#/components/schemas/main.problemJson' components: schemas: main.createMediaSearchConfigRequest: type: object properties: data: description: >- `data` contains the parameters for configuring a media search, such as keywords, risk categories, and language. type: object required: - keyword_lists - language - risk_categories properties: disambiguation_auto_risk_assessment: description: >- `disambiguation_auto_risk_assessment` if `active` is set `true`, articles with the set disambiguation `threshold` or below will be automatically risk-assessed as not risky. type: object properties: active: description: >- `active` indicates whether the `disambiguation_auto_risk_assessment`is active or not type: boolean example: true threshold: description: >- `threshold` if an article has disambiguation match the score of threshold or below, the article will be risk-assesses as not risky. type: integer example: 2 keyword_lists: description: >- `keyword_lists` contains a list of keyword list input items, each representing a group of keywords for the search configuration. type: array items: $ref: '#/components/schemas/main.NewKeyWordListInputItem' keywords: description: >- `WARNING: keywords are deprecated - use keyword_lists instead`. Your current implementation with `keywords` is still supported. type: array items: type: string example: - Fraud - Theft - Steal - Scam - Money Laundering - Sanction - Human Rights Violation - Terrorist - Organised Crime - Corrupt - Bribe - Tax Evasion - Convicted - Sentenced - Arrested - Child Labour language: description: >- `language` represents the language the search will be summarised in. The language and the language of the risk_categories need to match. type: string example: en look_back_period: description: >- `look_back_period` defines the look back period. Articles that got published before the look back period (today - days specified) will be ignored. allOf: - $ref: '#/components/schemas/main.lookBackPeriod' monitoring: description: >- `monitoring` defines the monitoring configuration details associated with the media search configuration request. allOf: - $ref: '#/components/schemas/main.monitoringInput' risk_categories: description: '`risk_categories` represents a list of risk types.' type: array items: type: string example: - Fraud - Money Laundering - Sanctions - Human Rights Abuse - Environmental Abuse - Terrorism - Organized Crime - Corruption - Tax Crimes - Violence - Theft - Discrimination - Other main.createSearchConfigResponse: type: object properties: data: type: object properties: media_search_configuration_id: description: >- `media_search_configuration_id` represents the unique identifier for a media search configuration. type: string example: 123e4567-e89b-12d3-a456-426614174000 main.problemJson: type: object properties: context: description: >- `context` provides additional information about the error, including the HTTP method and path where it occurred. allOf: - $ref: '#/components/schemas/main.defaultProblemContext' details: description: >- `details` provides a detailed human-readable explanation of the specific error. type: string example: Details explaining the error instance: description: >- `instance` specifies a unique URI identifying the context of the error, such as organisation, user, and request details. type: string example: /org/{org_id}/user/{user_id}/request-id/{request_id} title: description: >- `title` provides a short, human-readable summary of the problem type or error encountered. type: string example: Error ShortName type: description: >- `type` is a URI reference that identifies the problem type, allowing clients to recognize the type of error encountered. type: string example: https://errors.taidalos.com/common/{error-type} main.NewKeyWordListInputItem: type: object properties: keywords: type: array items: type: string example: - Fraud - Theft - Steal - Scam - Money Laundering - Sanction - Human Rights Violation - Terrorist - Organised Crime - Corrupt - Bribe - Tax Evasion - Convicted - Sentenced - Arrested - Child Labour name: type: string example: ESG Risk main.lookBackPeriod: type: object properties: days: description: >- `days` specifies the look back period in days, represented as an integer. All articles that got published before the look back period (today - days specified) will be ignored. type: integer example: 360 is_active: description: >- `is_active` indicates whether a look back period is enabled for the given configuration. type: boolean example: true main.monitoringInput: type: object properties: interval: description: >- `interval` specifies the frequency of monitoring, represented as a string. type: string example: WEEKLY is_active: description: >- `is_active` indicates whether monitoring is enabled for the given configuration. type: boolean example: true main.defaultProblemContext: type: object required: - method - path properties: method: description: >- `method` specifies the HTTP method (e.g., GET, POST) that was used in the problematic request. type: string example: '{method}' path: description: >- `path` specifies the HTTP request path associated with the encountered problem. type: string example: '{path}' timestamp: type: string ````