--- title: "Generate Document" method: POST path: "/interactions/{id}/documents/" tags: ["Documents (Classic)"] --- # Generate Document `POST /interactions/{id}/documents/` This endpoint offers different ways to generate a document. Find guides to document generation [here](/textgen/documents-standard). ## Path parameters - `id` string, uuid, required ## Headers - `Tenant-Name` string, required — Identifies a distinct entity within Corti's multi-tenant system. Ensures correct routing and authentication of the request. - `X-Corti-Retention-Policy` 'none' ## Request body - union - DocumentsCreateRequestWithTemplateKey - `context` DocumentsContext[], required — An array of context objects. Currently only accepts multiple objects when of type `transcript`. See [guide](/textgen/documents-standard#generate-document-from-transcript-as-input). - union - DocumentsContextWithFacts - `type` 'facts', required — The type of context data that will be used in the request: `Facts`, `Transcript`, or `String`. - `data` FactsContext[], required — An array of facts. See [guide](/textgen/documents-standard##generate-document-from-facts-as-input). - `text` string, required — The text of the fact. - `group` string — The group to which the fact belongs. - `source` 'core' | 'system' | 'user' - DocumentsContextWithTranscript - `type` 'transcript', required — The type of context data that will be used in the request: `Facts`, `Transcript`, or `String`. - `data` CommonTranscriptRequest, required - `channel` integer — The channel associated with this phrase/utterance. - `participant` integer — The identifier of the participant. - `speakerId` integer — Id to tag an identified speaker. Auto-increments. - `text` string, required — The spoken phrase or utterance extracted from the audio. - `start` integer — Start time in milliseconds for phrase/utterance. - `end` integer — End time in milliseconds for phrase/utterance. - DocumentsContextWithString - `type` 'string', required — The type of context data that will be used in the request: `Facts`, `Transcript`, or `String`. - `data` string, required — String data can include any text to be reasoned over for document generation: Transcript text, facts, or other narrative information. - `templateKey` string, required — The key of the template referencing the sections for generating a document. - `name` string — An optional name for the document. - `outputLanguage` string, required — The language in which the document will be generated. Check [languages page](/stt/languages) for more. - `disableGuardrails` boolean — Set to true to disable guardrails during document generation, default is false. - `documentationMode` 'global_sequential' | 'routed_parallel' — Configures the approach and underlying system prompt that govern how the LLM generates documentation. - DocumentsCreateRequestWithTemplate - `context` DocumentsContext[], required — An array of context objects. Currently only accepts multiple objects when of type `transcript`. See [guide](/textgen/documents-standard#generate-document-from-transcript-as-input). - union - DocumentsContextWithFacts - `type` 'facts', required — The type of context data that will be used in the request: `Facts`, `Transcript`, or `String`. - `data` FactsContext[], required — An array of facts. See [guide](/textgen/documents-standard##generate-document-from-facts-as-input). - `text` string, required — The text of the fact. - `group` string — The group to which the fact belongs. - `source` 'core' | 'system' | 'user' - DocumentsContextWithTranscript - `type` 'transcript', required — The type of context data that will be used in the request: `Facts`, `Transcript`, or `String`. - `data` CommonTranscriptRequest, required - `channel` integer — The channel associated with this phrase/utterance. - `participant` integer — The identifier of the participant. - `speakerId` integer — Id to tag an identified speaker. Auto-increments. - `text` string, required — The spoken phrase or utterance extracted from the audio. - `start` integer — Start time in milliseconds for phrase/utterance. - `end` integer — End time in milliseconds for phrase/utterance. - DocumentsContextWithString - `type` 'string', required — The type of context data that will be used in the request: `Facts`, `Transcript`, or `String`. - `data` string, required — String data can include any text to be reasoned over for document generation: Transcript text, facts, or other narrative information. - `template` union, required - DocumentsTemplateWithSections - `sections` DocumentsSectionOverride[], required - `key` string, required — The key that references the section to use for document generation. - `nameOverride` string — Overrides the section name used in document generation and response. - `writingStyleOverride` string — Overrides the section's default writing style with your custom prompt. - `formatRuleOverride` string — Overrides the section's default format rule with your custom prompt. - `additionalInstructionsOverride` string — Overrides and sets the section-level additional instructions with your custom prompt. - `contentOverride` string — Overrides the section's content prompt used for input assignment with documentationMode: routed_parallel, and section generation. - `description` string — A brief description of the document that can help give the LLM some context. - `additionalInstructionsOverride` string — Overrides and sets template-level additional instructions. - DocumentsTemplateWithSectionKeys - `sectionKeys` string[], required — An array of section keys. - `documentName` string — The name of the document. - `additionalInstructions` string — Any additional instructions to be considered during document generation. - `name` string — An optional name for the document. - `outputLanguage` string, required — The language in which the document will be generated. Check [languages page](/stt/languages) for more. - `disableGuardrails` boolean — Set to true to disable guardrails during document generation, default is false. - `documentationMode` 'global_sequential' | 'routed_parallel' — Configures the approach and underlying system prompt that govern how the LLM generates documentation. ## Response `201` - DocumentsGetResponse - `id` string, uuid, required - `name` string, required — Name of the generated document - `templateRef` string, required — Reference for the used template - `isStream` boolean, required - `sections` DocumentsSection[], required — Individual document sections - `key` string, required — Document section key - `name` string, required — Name or heading of the document section within the generated document - `text` string, required — Contents of the document section within the generated document - `sort` integer, required — Order of the document section within the generated document - `createdAt` string, date-time, required — The original timestamp when the document section was created. - `updatedAt` string, date-time, required — The timestamp when the document section was last updated. - `createdAt` string, date-time, required — The original timestamp when the document was created. - `updatedAt` string, date-time, required — The timestamp when the document was last updated. - `outputLanguage` string, required — The language in which the document will be generated. Check https://docs.corti.ai/stt/languages for more. - `usageInfo` CommonUsageInfo, required — Credits consumed for this request. - `creditsConsumed` number, required ## Other responses - `400` — RFC9457 - `403` — RFC9457 - `500` — RFC9457 - `504` — RFC9457 --- [API](https://skmtc.net/corti/apis/admin-api.md) · [All operations](https://skmtc.net/corti/apis/admin-api/llms.txt) · [OpenAPI document](https://skmtc-service-staging.skmtc.workers.dev/v1/apis/corti/admin-api/revisions/5d1895a1b3fc/schema)